wire to grpc

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

.devcontainer-scripts/postcreate.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+
+cd /workspace
+
+# Get the files into the volume without a bind mount
+if [ ! -d ".git" ]; then
+    git clone https://github.com/mudler/LocalAI.git .
+else
+    git fetch
+fi
+
+echo "Standard Post-Create script completed."
+
+if [ -f "/devcontainer-customization/postcreate.sh" ]; then
+    echo "Launching customization postcreate.sh"
+    bash "/devcontainer-customization/postcreate.sh"
+fi

.devcontainer-scripts/poststart.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+
+cd /workspace
+
+# Grab the pre-stashed backend assets to avoid build issues
+cp -r /build/backend-assets /workspace/backend-assets
+
+# Ensures generated source files are present upon load
+make prepare
+
+echo "Standard Post-Start script completed."
+
+if [ -f "/devcontainer-customization/poststart.sh" ]; then
+    echo "Launching customization poststart.sh"
+    bash "/devcontainer-customization/poststart.sh"
+fi

.devcontainer-scripts/utils.sh

@@ -0,0 +1,55 @@
+#!/bin/bash
+
+# This file contains some really simple functions that are useful when building up customization scripts.
+
+
+# Checks if the git config has a user registered - and sets it up if not.
+#
+# Param 1: name
+# Param 2: email
+#
+config_user() {
+    echo "Configuring git for $1 <$2>"
+    local gcn=$(git config --global user.name)
+    if [ -z "${gcn}" ]; then
+        echo "Setting up git user / remote"
+        git config --global user.name "$1"
+        git config --global user.email "$2"
+        
+    fi
+}
+
+# Checks if the git remote is configured - and sets it up if not. Fetches either way.
+#
+# Param 1: remote name
+# Param 2: remote url
+#
+config_remote() {
+    echo "Adding git remote and fetching $2 as $1"
+    local gr=$(git remote -v | grep $1)
+    if [ -z "${gr}" ]; then
+        git remote add $1 $2
+    fi
+    git fetch $1
+}
+
+# Setup special .ssh files
+# Prints out lines of text to make things pretty
+# Param 1: bash array, filenames relative to the customization directory that should be copied to ~/.ssh
+setup_ssh() {
+    echo "starting ~/.ssh directory setup..."
+    mkdir -p "${HOME}.ssh"
+    chmod 0700 "${HOME}/.ssh"
+    echo "-----"
+    local files=("$@")
+    for file in "${files[@]}" ; do
+        local cfile="/devcontainer-customization/${file}"
+        local hfile="${HOME}/.ssh/${file}"
+        if [ ! -f "${hfile}" ]; then
+            echo "copying \"${file}\""
+            cp "${cfile}" "${hfile}"
+            chmod 600 "${hfile}"
+        fi
+    done
+    echo "~/.ssh directory setup complete!"
+}

.devcontainer/Dockerfile

@@ -1,3 +0,0 @@
-ARG GO_VERSION=1.20
-FROM mcr.microsoft.com/devcontainers/go:0-$GO_VERSION-bullseye
-RUN apt-get update && apt-get install -y cmake

.devcontainer/customization/README.md

@@ -0,0 +1,25 @@
+Place any additional resources your environment requires in this directory
+
+Script hooks are currently called for:
+`postcreate.sh` and `poststart.sh`
+
+If files with those names exist here, they will be called at the end of the normal script.
+
+This is a good place to set things like `git config --global user.name` are set - and to handle any other files that are mounted via this directory.
+
+To assist in doing so, `source /.devcontainer-scripts/utils.sh` will provide utility functions that may be useful - for example:
+
+```
+#!/bin/bash
+
+source "/.devcontainer-scripts/utils.sh"
+
+sshfiles=("config", "key.pub")
+
+setup_ssh "${sshfiles[@]}"
+
+config_user "YOUR NAME" "YOUR EMAIL"
+
+config_remote "REMOTE NAME" "REMOTE URL"
+
+```

.devcontainer/devcontainer.json

@@ -1,46 +1,24 @@
-// For format details, see https://aka.ms/devcontainer.json. For config options, see the
-// README at: https://github.com/devcontainers/templates/tree/main/src/docker-existing-docker-compose
-{
-	"name": "Existing Docker Compose (Extend)",
-
-	// Update the 'dockerComposeFile' list if you have more compose files or use different names.
-	// The .devcontainer/docker-compose.yml file contains any overrides you need/want to make.
-	"dockerComposeFile": [
-		"../docker-compose.yaml",
-		"docker-compose.yml"
-	],
-
-	// The 'service' property is the name of the service for the container that VS Code should
-	// use. Update this value and .devcontainer/docker-compose.yml to the real service name.
-	"service": "api",
-
-	// The optional 'workspaceFolder' property is the path VS Code should open by default when
-	// connected. This is typically a file mount in .devcontainer/docker-compose.yml
-	"workspaceFolder": "/workspace",
-
-	"features": {
-		"ghcr.io/devcontainers/features/go:1": {},
-		"ghcr.io/azutake/devcontainer-features/go-packages-install:0": {}
-	},
-
-	// Features to add to the dev container. More info: https://containers.dev/features.
-	// "features": {},
-
-	// Use 'forwardPorts' to make a list of ports inside the container available locally.
-	// "forwardPorts": [],
-
-	// Uncomment the next line if you want start specific services in your Docker Compose config.
-	// "runServices": [],
-
-	// Uncomment the next line if you want to keep your containers running after VS Code shuts down.
-	// "shutdownAction": "none",
-
-	// Uncomment the next line to run commands after the container is created.
-	"postCreateCommand": "make prepare"
-
-	// Configure tool-specific properties.
-	// "customizations": {},
-
-	// Uncomment to connect as an existing user other than the container default. More info: https://aka.ms/dev-containers-non-root.
-	// "remoteUser": "devcontainer"
-}
+{
+    "$schema": "https://raw.githubusercontent.com/devcontainers/spec/main/schemas/devContainer.schema.json",
+    "name": "LocalAI",
+    "workspaceFolder": "/workspace",
+    "dockerComposeFile": [ "./docker-compose-devcontainer.yml" ],
+    "service": "api",
+    "shutdownAction": "stopCompose",
+    "customizations": {
+        "vscode": {
+            "extensions": [
+                "golang.go",
+                "ms-vscode.makefile-tools",
+                "ms-azuretools.vscode-docker",
+                "ms-python.python",
+                "ms-python.debugpy",
+                "wayou.vscode-todo-highlight",
+                "waderyan.gitblame"
+            ]
+        }
+    },
+    "forwardPorts": [8080, 3000],
+    "postCreateCommand": "bash /.devcontainer-scripts/postcreate.sh",
+    "postStartCommand": "bash /.devcontainer-scripts/poststart.sh"
+}

.devcontainer/docker-compose-devcontainer.yml

@@ -0,0 +1,48 @@
+services:
+  api:
+    build:
+      context: ..
+      dockerfile: Dockerfile
+      target: devcontainer
+      args:
+      - FFMPEG=true
+      - IMAGE_TYPE=extras
+      - GO_TAGS=p2p tts
+    env_file:
+      - ../.env
+    ports:
+      - 8080:8080
+    volumes:
+      - localai_workspace:/workspace
+      - ../models:/host-models
+      - ./customization:/devcontainer-customization
+    command: /bin/sh -c "while sleep 1000; do :; done"
+    cap_add:
+      - SYS_PTRACE
+    security_opt:
+      - seccomp:unconfined
+  prometheus:
+    image: prom/prometheus
+    container_name: prometheus
+    command:
+      - '--config.file=/etc/prometheus/prometheus.yml'
+    ports:
+      - 9090:9090
+    restart: unless-stopped
+    volumes:
+      - ./prometheus:/etc/prometheus
+      - prom_data:/prometheus
+  grafana:
+    image: grafana/grafana
+    container_name: grafana
+    ports:
+      - 3000:3000
+    restart: unless-stopped
+    environment:
+      - GF_SECURITY_ADMIN_USER=admin
+      - GF_SECURITY_ADMIN_PASSWORD=grafana
+    volumes:
+      - ./grafana:/etc/grafana/provisioning/datasources
+volumes:
+  prom_data:
+  localai_workspace:

.devcontainer/docker-compose.yml

@@ -1,26 +0,0 @@
-version: '3.6'
-services:
-  # Update this to the name of the service you want to work with in your docker-compose.yml file
-  api:
-    # Uncomment if you want to override the service's Dockerfile to one in the .devcontainer 
-    # folder. Note that the path of the Dockerfile and context is relative to the *primary* 
-    # docker-compose.yml file (the first in the devcontainer.json "dockerComposeFile"
-    # array). The sample below assumes your primary file is in the root of your project.
-    #
-    build:
-      context: .
-      dockerfile: .devcontainer/Dockerfile
-
-    volumes:
-      # Update this to wherever you want VS Code to mount the folder of your project
-      - .:/workspace:cached
-
-    # Uncomment the next four lines if you will use a ptrace-based debugger like C++, Go, and Rust.
-    # cap_add:
-    #   - SYS_PTRACE
-    # security_opt:
-    #   - seccomp:unconfined
-
-    # Overrides default command so things don't shut down after the process ends.
-    command: /bin/sh -c "while sleep 1000; do :; done"
- 

.devcontainer/grafana/datasource.yml

@@ -0,0 +1,10 @@
+
+apiVersion: 1
+
+datasources:
+- name: Prometheus
+  type: prometheus
+  url: http://prometheus:9090 
+  isDefault: true
+  access: proxy
+  editable: true

.devcontainer/prometheus/prometheus.yml

@@ -0,0 +1,21 @@
+global:
+  scrape_interval: 15s
+  scrape_timeout: 10s
+  evaluation_interval: 15s
+alerting:
+  alertmanagers:
+  - static_configs:
+    - targets: []
+    scheme: http
+    timeout: 10s
+    api_version: v1
+scrape_configs:
+- job_name: prometheus
+  honor_timestamps: true
+  scrape_interval: 15s
+  scrape_timeout: 10s
+  metrics_path: /metrics
+  scheme: http
+  static_configs:
+  - targets:
+    - localhost:9090

.dockerignore

@@ -1,4 +1,17 @@
+.idea
+.github
+.vscode
+.devcontainer
 models
 examples/chatbot-ui/models
 examples/rwkv/models
 examples/**/models
+Dockerfile*
+__pycache__
+
+# SonarQube
+.scannerwork
+
+# backend virtual environments
+**/venv
+backend/python/**/source

.editorconfig

@@ -0,0 +1,31 @@
+
+root = true
+
+[*]
+indent_style = space
+indent_size = 2
+end_of_line = lf
+charset = utf-8
+trim_trailing_whitespace = true
+insert_final_newline = true
+
+[*.go]
+indent_style = tab
+
+[Makefile]
+indent_style = tab
+
+[*.proto]
+indent_size = 2
+
+[*.py]
+indent_size = 4
+
+[*.js]
+indent_size = 2
+
+[*.yaml]
+indent_size = 2
+
+[*.md]
+trim_trailing_whitespace = false

.env

@@ -1,5 +1,109 @@
-# THREADS=14
-# CONTEXT_SIZE=512
-MODELS_PATH=/models
-# DEBUG=true
-# BUILD_TYPE=generic
+## Set number of threads.
+## Note: prefer the number of physical cores. Overbooking the CPU degrades performance notably.
+# LOCALAI_THREADS=14
+
+## Specify a different bind address (defaults to ":8080")
+# LOCALAI_ADDRESS=127.0.0.1:8080
+
+## Default models context size
+# LOCALAI_CONTEXT_SIZE=512
+#
+## Define galleries.
+## models will to install will be visible in `/models/available`
+# LOCALAI_GALLERIES=[{"name":"localai", "url":"github:mudler/LocalAI/gallery/index.yaml@master"}]
+
+## CORS settings
+# LOCALAI_CORS=true
+# LOCALAI_CORS_ALLOW_ORIGINS=*
+
+## Default path for models
+#
+# LOCALAI_MODELS_PATH=/models
+
+## Enable debug mode
+# LOCALAI_LOG_LEVEL=debug
+
+## Disables COMPEL (Diffusers)
+# COMPEL=0
+
+## Enable/Disable single backend (useful if only one GPU is available)
+# LOCALAI_SINGLE_ACTIVE_BACKEND=true
+
+# Forces shutdown of the backends if busy (only if LOCALAI_SINGLE_ACTIVE_BACKEND is set)
+# LOCALAI_FORCE_BACKEND_SHUTDOWN=true
+
+## Specify a build type. Available: cublas, openblas, clblas.
+## cuBLAS: This is a GPU-accelerated version of the complete standard BLAS (Basic Linear Algebra Subprograms) library. It's provided by Nvidia and is part of their CUDA toolkit.
+## OpenBLAS: This is an open-source implementation of the BLAS library that aims to provide highly optimized code for various platforms. It includes support for multi-threading and can be compiled to use hardware-specific features for additional performance. OpenBLAS can run on many kinds of hardware, including CPUs from Intel, AMD, and ARM.
+## clBLAS:   This is an open-source implementation of the BLAS library that uses OpenCL, a framework for writing programs that execute across heterogeneous platforms consisting of CPUs, GPUs, and other processors. clBLAS is designed to take advantage of the parallel computing power of GPUs but can also run on any hardware that supports OpenCL. This includes hardware from different vendors like Nvidia, AMD, and Intel.
+# BUILD_TYPE=openblas
+
+## Uncomment and set to true to enable rebuilding from source
+# REBUILD=true
+
+## Enable go tags, available: p2p, tts
+## p2p: enable distributed inferencing
+## tts: enables text-to-speech with go-piper 
+## (requires REBUILD=true)
+#
+# GO_TAGS=p2p
+
+## Path where to store generated images
+# LOCALAI_IMAGE_PATH=/tmp/generated/images
+
+## Specify a default upload limit in MB (whisper)
+# LOCALAI_UPLOAD_LIMIT=15
+
+## List of external GRPC backends (note on the container image this variable is already set to use extra backends available in extra/)
+# LOCALAI_EXTERNAL_GRPC_BACKENDS=my-backend:127.0.0.1:9000,my-backend2:/usr/bin/backend.py
+
+### Advanced settings ###
+### Those are not really used by LocalAI, but from components in the stack ###
+##
+### Preload libraries
+# LD_PRELOAD=
+
+### Huggingface cache for models
+# HUGGINGFACE_HUB_CACHE=/usr/local/huggingface
+
+### Python backends GRPC max workers
+### Default number of workers for GRPC Python backends.
+### This actually controls wether a backend can process multiple requests or not.
+# PYTHON_GRPC_MAX_WORKERS=1
+
+### Define the number of parallel LLAMA.cpp workers (Defaults to 1)
+# LLAMACPP_PARALLEL=1
+
+### Define a list of GRPC Servers for llama-cpp workers to distribute the load
+# https://github.com/ggerganov/llama.cpp/pull/6829
+# https://github.com/ggerganov/llama.cpp/blob/master/examples/rpc/README.md
+# LLAMACPP_GRPC_SERVERS=""
+
+### Enable to run parallel requests
+# LOCALAI_PARALLEL_REQUESTS=true
+
+# Enable to allow p2p mode
+# LOCALAI_P2P=true
+
+# Enable to use federated mode
+# LOCALAI_FEDERATED=true
+
+# Enable to start federation server
+# FEDERATED_SERVER=true
+
+# Define to use federation token
+# TOKEN=""
+
+### Watchdog settings
+###
+# Enables watchdog to kill backends that are inactive for too much time
+# LOCALAI_WATCHDOG_IDLE=true
+#
+# Time in duration format (e.g. 1h30m) after which a backend is considered idle
+# LOCALAI_WATCHDOG_IDLE_TIMEOUT=5m
+#
+# Enables watchdog to kill backends that are busy for too much time
+# LOCALAI_WATCHDOG_BUSY=true
+#
+# Time in duration format (e.g. 1h30m) after which a backend is considered busy
+# LOCALAI_WATCHDOG_BUSY_TIMEOUT=5m

.gitattributes

@@ -0,0 +1,2 @@
+*.sh text eol=lf
+backend/cpp/llama/*.hpp linguist-vendored

.github/FUNDING.yml

@@ -0,0 +1,5 @@
+# These are supported funding model platforms
+
+github: [mudler]
+custom: 
+- https://www.buymeacoffee.com/mudler

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,29 @@
+---
+name: Bug report
+about: Create a report to help us improve
+title: ''
+labels: bug, unconfirmed, up-for-grabs
+---
+
+<!-- Thanks for helping us to improve LocalAI! We welcome all bug reports. Please fill out each area of the template so we can better help you. Comments like this will be hidden when you post but you can delete them if you wish. -->
+
+**LocalAI version:**
+<!-- Container Image or LocalAI tag/commit -->
+
+**Environment, CPU architecture, OS, and Version:**
+<!-- Provide the output from "uname -a", HW specs, if it's a VM  -->
+
+**Describe the bug**
+<!-- A clear and concise description of what the bug is. -->
+
+**To Reproduce**
+<!-- Steps to reproduce the behavior, including the LocalAI command used, if any -->
+
+**Expected behavior**
+<!-- A clear and concise description of what you expected to happen. -->
+
+**Logs**
+<!-- If applicable, add logs while running LocalAI in debug mode (`--debug` or `DEBUG=true`) to help explain your problem.  -->
+
+**Additional context**
+<!-- Add any other context about the problem here. -->

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,8 @@
+blank_issues_enabled: false
+contact_links:
+  - name: Community Support
+    url: https://github.com/go-skynet/LocalAI/discussions
+    about: Please ask and answer questions here.
+  - name: Discord
+    url: https://discord.gg/uJAeKSAGDy
+    about: Join our community on Discord!

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,20 @@
+---
+name: Feature request
+about: Suggest an idea for this project
+title: ''
+labels: enhancement, up-for-grabs
+---
+
+<!-- Thanks for helping us to improve LocalAI! We welcome all feature requests. Please fill out each area of the template so we can better help you. Comments like this will be hidden when you post but you can delete them if you wish. -->
+
+**Is your feature request related to a problem? Please describe.**
+<!-- A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]  -->
+
+**Describe the solution you'd like**
+<!-- A clear and concise description of what you want to happen.  -->
+
+**Describe alternatives you've considered**
+<!-- A clear and concise description of any alternative solutions or features you've considered.  -->
+
+**Additional context**
+<!-- Add any other context or screenshots about the feature request here. -->

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,31 @@
+**Description**
+
+This PR fixes #
+
+**Notes for Reviewers**
+
+
+**[Signed commits](../CONTRIBUTING.md#signing-off-on-commits-developer-certificate-of-origin)**
+- [ ] Yes, I signed my commits.
+ 
+<!--
+Thank you for contributing to LocalAI! 
+
+Contributing Conventions
+-------------------------
+
+The draft above helps to give a quick overview of your PR.
+
+Remember to remove this comment and to at least:
+
+1. Include descriptive PR titles with [<component-name>] prepended. We use [conventional commits](https://www.conventionalcommits.org/en/v1.0.0/).
+2. Build and test your changes before submitting a PR (`make build`). 
+3. Sign your commits
+4. **Tag maintainer:** for a quicker response, tag the relevant maintainer (see below).
+5. **X/Twitter handle:** we announce bigger features on X/Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out!
+
+By following the community's contribution conventions upfront, the review process will 
+be accelerated and your PR merged more quickly.
+
+If no one reviews your PR within a few days, please @-mention @mudler.
+-->

.github/bump_deps.sh

@@ -6,4 +6,17 @@ VAR=$3
 
 LAST_COMMIT=$(curl -s -H "Accept: application/vnd.github.VERSION.sha" "https://api.github.com/repos/$REPO/commits/$BRANCH")
 
+# Read $VAR from Makefile (only first match)
+set +e
+CURRENT_COMMIT="$(grep -m1 "^$VAR?=" Makefile | cut -d'=' -f2)"
+set -e
+
 sed -i Makefile -e "s/$VAR?=.*/$VAR?=$LAST_COMMIT/"
+
+if [ -z "$CURRENT_COMMIT" ]; then
+    echo "Could not find $VAR in Makefile."
+    exit 0
+fi
+
+echo "Changes: https://github.com/$REPO/compare/${CURRENT_COMMIT}..${LAST_COMMIT}" >> "${VAR}_message.txt"
+echo "${LAST_COMMIT}" >> "${VAR}_commit.txt"

.github/bump_docs.sh

@@ -0,0 +1,7 @@
+#!/bin/bash
+set -xe
+REPO=$1
+
+LATEST_TAG=$(curl -s "https://api.github.com/repos/$REPO/releases/latest" | jq -r '.tag_name')
+
+cat <<< $(jq ".version = \"$LATEST_TAG\"" docs/data/version.json) > docs/data/version.json

.github/check_and_update.py

@@ -0,0 +1,85 @@
+import hashlib
+from huggingface_hub import hf_hub_download, get_paths_info
+import requests
+import sys
+import os
+
+uri = sys.argv[1]
+file_name = uri.split('/')[-1]
+
+# Function to parse the URI and determine download method
+def parse_uri(uri):
+    if uri.startswith('huggingface://'):
+        repo_id = uri.split('://')[1]
+        return 'huggingface', repo_id.rsplit('/', 1)[0]
+    elif 'huggingface.co' in uri:
+        parts = uri.split('/resolve/')
+        if len(parts) > 1:
+            repo_path = parts[0].split('https://huggingface.co/')[-1]
+            return 'huggingface', repo_path
+    return 'direct', uri
+
+def calculate_sha256(file_path):
+    sha256_hash = hashlib.sha256()
+    with open(file_path, 'rb') as f:
+        for byte_block in iter(lambda: f.read(4096), b''):
+            sha256_hash.update(byte_block)
+    return sha256_hash.hexdigest()
+
+def manual_safety_check_hf(repo_id):
+    scanResponse = requests.get('https://huggingface.co/api/models/' + repo_id + "/scan")
+    scan = scanResponse.json()
+    # Check if 'hasUnsafeFile' exists in the response
+    if 'hasUnsafeFile' in scan:
+        if scan['hasUnsafeFile']:
+            return scan
+        else:
+            return None
+    else:
+        return None
+
+download_type, repo_id_or_url = parse_uri(uri)
+
+new_checksum =  None
+file_path = None
+
+# Decide download method based on URI type
+if download_type == 'huggingface':
+    # Check if the repo is flagged as dangerous by HF
+    hazard = manual_safety_check_hf(repo_id_or_url)
+    if hazard != None:
+        print(f'Error: HuggingFace has detected security problems for {repo_id_or_url}: {str(hazard)}', filename=file_name)
+        sys.exit(5)
+    # Use HF API to pull sha
+    for file in get_paths_info(repo_id_or_url, [file_name], repo_type='model'):
+        try:
+            new_checksum = file.lfs.sha256
+            break
+        except Exception as e:
+            print(f'Error from Hugging Face Hub: {str(e)}', file=sys.stderr)
+            sys.exit(2)
+    if new_checksum is None:
+        try:
+            file_path = hf_hub_download(repo_id=repo_id_or_url, filename=file_name)
+        except Exception as e:
+            print(f'Error from Hugging Face Hub: {str(e)}', file=sys.stderr)
+            sys.exit(2)
+else:
+    response = requests.get(repo_id_or_url)
+    if response.status_code == 200:
+        with open(file_name, 'wb') as f:
+            f.write(response.content)
+        file_path = file_name
+    elif response.status_code == 404:
+        print(f'File not found: {response.status_code}', file=sys.stderr)
+        sys.exit(2)
+    else:
+        print(f'Error downloading file: {response.status_code}', file=sys.stderr)
+        sys.exit(1)
+
+if new_checksum is None:
+    new_checksum = calculate_sha256(file_path)
+    print(new_checksum)
+    os.remove(file_path)
+else:
+    print(new_checksum)

.github/checksum_checker.sh

@@ -0,0 +1,63 @@
+#!/bin/bash
+# This scripts needs yq and huggingface_hub to be installed
+# to install hugingface_hub run pip install huggingface_hub
+
+# Path to the input YAML file
+input_yaml=$1
+
+# Function to download file and check checksum using Python
+function check_and_update_checksum() {
+    model_name="$1"
+    file_name="$2"
+    uri="$3"
+    old_checksum="$4"
+    idx="$5"
+
+    # Download the file and calculate new checksum using Python
+    new_checksum=$(python3 ./.github/check_and_update.py $uri)
+    result=$?
+
+    if [[ $result -eq 5 ]]; then
+        echo "Contaminated entry detected, deleting entry for $model_name..."
+        yq eval -i "del([$idx])" "$input_yaml"
+        return
+    fi
+
+    if [[ "$new_checksum" == "" ]]; then
+        echo "Error calculating checksum for $file_name. Skipping..."
+        return
+    fi
+
+    echo "Checksum for $file_name: $new_checksum"
+
+    # Compare and update the YAML file if checksums do not match
+    
+    if [[ $result -eq 2 ]]; then
+        echo "File not found, deleting entry for $file_name..."
+        # yq eval -i "del(.[$idx].files[] | select(.filename == \"$file_name\"))" "$input_yaml"
+    elif [[ "$old_checksum" != "$new_checksum" ]]; then
+        echo "Checksum mismatch for $file_name. Updating..."
+        yq eval -i "del(.[$idx].files[] | select(.filename == \"$file_name\").sha256)" "$input_yaml"
+        yq eval -i "(.[$idx].files[] | select(.filename == \"$file_name\")).sha256 = \"$new_checksum\"" "$input_yaml"
+    elif [[ $result -ne 0 ]]; then
+        echo "Error downloading file $file_name. Skipping..."
+    else
+        echo "Checksum match for $file_name. No update needed."
+    fi
+}
+
+# Read the YAML and process each file
+len=$(yq eval '. | length' "$input_yaml")
+for ((i=0; i<$len; i++))
+do
+    name=$(yq eval ".[$i].name" "$input_yaml")
+    files_len=$(yq eval ".[$i].files | length" "$input_yaml")
+    for ((j=0; j<$files_len; j++))
+    do
+        filename=$(yq eval ".[$i].files[$j].filename" "$input_yaml")
+        uri=$(yq eval ".[$i].files[$j].uri" "$input_yaml")
+        checksum=$(yq eval ".[$i].files[$j].sha256" "$input_yaml")
+        echo "Checking model $name, file $filename. URI = $uri, Checksum = $checksum"
+        check_and_update_checksum "$name" "$filename" "$uri" "$checksum" "$i"
+    done
+done

.github/ci/modelslist.go

@@ -0,0 +1,304 @@
+package main
+
+import (
+	"fmt"
+	"html/template"
+	"io/ioutil"
+	"os"
+
+	"github.com/microcosm-cc/bluemonday"
+	"gopkg.in/yaml.v3"
+)
+
+var modelPageTemplate string = `
+<!DOCTYPE html>
+<html>
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>LocalAI models</title>
+    <link href="https://cdnjs.cloudflare.com/ajax/libs/flowbite/2.3.0/flowbite.min.css" rel="stylesheet" />
+    <script src="https://cdn.jsdelivr.net/npm/vanilla-lazyload@19.1.3/dist/lazyload.min.js"></script>
+
+    <link
+    rel="stylesheet"
+    href="https://cdn.jsdelivr.net/gh/highlightjs/cdn-release@11.8.0/build/styles/default.min.css"
+  />
+    <script
+    defer
+    src="https://cdn.jsdelivr.net/gh/highlightjs/cdn-release@11.8.0/build/highlight.min.js"
+  ></script>
+    <script
+    defer
+    src="https://cdn.jsdelivr.net/npm/alpinejs@3.x.x/dist/cdn.min.js"
+  ></script>
+  <script
+    defer
+    src="https://cdn.jsdelivr.net/npm/marked/marked.min.js"
+  ></script>
+  <script
+    defer
+    src="https://cdn.jsdelivr.net/npm/dompurify@3.0.6/dist/purify.min.js"
+  ></script>
+
+  <link href="/static/general.css" rel="stylesheet" />
+    <link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;600;700&family=Roboto:wght@400;500&display=swap" rel="stylesheet">
+    <link
+    href="https://fonts.googleapis.com/css?family=Roboto:300,400,500,700,900&display=swap"
+    rel="stylesheet" />
+  <link
+    rel="stylesheet"
+    href="https://cdn.jsdelivr.net/npm/tw-elements/css/tw-elements.min.css" />
+  <script src="https://cdn.tailwindcss.com/3.3.0"></script>
+  <script>
+    tailwind.config = {
+      darkMode: "class",
+      theme: {
+        fontFamily: {
+          sans: ["Roboto", "sans-serif"],
+          body: ["Roboto", "sans-serif"],
+          mono: ["ui-monospace", "monospace"],
+        },
+      },
+      corePlugins: {
+        preflight: false,
+      },
+    };
+  </script>
+    <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.1.1/css/all.min.css">
+    <script src="https://unpkg.com/htmx.org@1.9.12" integrity="sha384-ujb1lZYygJmzgSwoxRggbCHcjc0rB2XoQrxeTUQyRjrOnlCoYta87iKBWq3EsdM2" crossorigin="anonymous"></script>
+</head>
+
+<body class="bg-gray-900 text-gray-200">
+<div class="flex flex-col min-h-screen">
+
+<nav class="bg-gray-800 shadow-lg">
+    <div class="container mx-auto px-4 py-4">
+        <div class="flex items-center justify-between">
+            <div class="flex items-center">
+                <a href="/" class="text-white text-xl font-bold"><img src="https://github.com/mudler/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo" class="h-10 mr-3 border-2 border-gray-300 shadow rounded"></a>
+                <a href="/" class="text-white text-xl font-bold">LocalAI</a>
+            </div>
+            <!-- Menu button for small screens -->
+            <div class="lg:hidden">
+                <button id="menu-toggle" class="text-gray-400 hover:text-white focus:outline-none">
+                    <i class="fas fa-bars fa-lg"></i>
+                </button>
+            </div>
+            <!-- Navigation links -->
+            <div class="hidden lg:flex lg:items-center lg:justify-end lg:flex-1 lg:w-0">
+                <a href="https://localai.io" class="text-gray-400 hover:text-white px-3 py-2 rounded" target="_blank" ><i class="fas fa-book-reader pr-2"></i> Documentation</a>
+            </div>
+        </div>
+        <!-- Collapsible menu for small screens -->
+        <div class="hidden lg:hidden" id="mobile-menu">
+            <div class="pt-4 pb-3 border-t border-gray-700">
+
+                <a href="https://localai.io" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1" target="_blank" ><i class="fas fa-book-reader pr-2"></i> Documentation</a>
+
+            </div>
+        </div>
+    </div>
+</nav>
+
+<style>
+  .is-hidden {
+	display: none;
+	  }
+</style>
+
+<div class="container mx-auto px-4 flex-grow">
+
+<div class="models mt-12">
+	<h2 class="text-center text-3xl font-semibold text-gray-100">
+	LocalAI model gallery list </h2><br>
+
+	<h2 class="text-center text-3xl font-semibold text-gray-100">
+
+	 🖼️ Available {{.AvailableModels}} models</i> <a href="https://localai.io/models/" target="_blank" >
+			<i class="fas fa-circle-info pr-2"></i>
+		</a></h2>
+
+	<h3>
+	Refer to the Model gallery <a href="https://localai.io/models/" target="_blank" ><i class="fas fa-circle-info pr-2"></i></a> for more information on how to use the models with LocalAI.<br>
+
+	You can install models with the CLI command <code>local-ai models install <model-name></code>. or by using the WebUI.
+	</h3>
+
+	<input class="form-control appearance-none block w-full mt-5 px-3 py-2 text-base font-normal text-gray-300 pb-2 mb-5 bg-gray-800 bg-clip-padding border border-solid border-gray-600 rounded transition ease-in-out m-0 focus:text-gray-300 focus:bg-gray-900 focus:border-blue-500 focus:outline-none" type="search"
+	id="searchbox" placeholder="Live search keyword..">
+	  <div class="dark grid grid-cols-1 grid-rows-1 md:grid-cols-3 block rounded-lg shadow-secondary-1 dark:bg-surface-dark">
+		{{ range $_, $model := .Models }}
+		<div class="box me-4 mb-2 block rounded-lg bg-white shadow-secondary-1  dark:bg-gray-800 dark:bg-surface-dark dark:text-white text-surface pb-2">
+		<div>
+		    {{ $icon := "https://upload.wikimedia.org/wikipedia/commons/6/65/No-Image-Placeholder.svg" }}
+			{{ if $model.Icon }}
+	  		{{ $icon = $model.Icon }}
+	  		{{ end }}
+			<div class="flex justify-center items-center">
+				<img data-src="{{ $icon }}" alt="{{$model.Name}}" class="rounded-t-lg max-h-48 max-w-96 object-cover mt-3 lazy">
+			</div>
+	  		<div class="p-6 text-surface dark:text-white">
+				<h5 class="mb-2 text-xl font-medium leading-tight">{{$model.Name}}</h5>
+
+
+				<p class="mb-4 text-base truncate">{{ $model.Description }}</p>
+
+			</div>
+			<div class="px-6 pt-4 pb-2">
+
+      <!-- Modal toggle -->
+      <button data-modal-target="{{ $model.Name}}-modal" data-modal-toggle="{{ $model.Name }}-modal" class="block text-white bg-blue-700 hover:bg-blue-800 focus:ring-4 focus:outline-none focus:ring-blue-300 font-medium rounded-lg text-sm px-5 py-2.5 text-center dark:bg-blue-600 dark:hover:bg-blue-700 dark:focus:ring-blue-800" type="button">
+        More info
+      </button>
+
+    <!-- Main modal -->
+    <div id="{{ $model.Name}}-modal" tabindex="-1" aria-hidden="true" class="hidden overflow-y-auto overflow-x-hidden fixed top-0 right-0 left-0 z-50 justify-center items-center w-full md:inset-0 h-[calc(100%-1rem)] max-h-full">
+        <div class="relative p-4 w-full max-w-2xl max-h-full">
+            <!-- Modal content -->
+            <div class="relative bg-white rounded-lg shadow dark:bg-gray-700">
+                <!-- Modal header -->
+                <div class="flex items-center justify-between p-4 md:p-5 border-b rounded-t dark:border-gray-600">
+                    <h3 class="text-xl font-semibold text-gray-900 dark:text-white">
+                        {{ $model.Name}}
+                    </h3>
+                    <button type="button" class="text-gray-400 bg-transparent hover:bg-gray-200 hover:text-gray-900 rounded-lg text-sm w-8 h-8 ms-auto inline-flex justify-center items-center dark:hover:bg-gray-600 dark:hover:text-white" data-modal-hide="{{$model.Name}}-modal">
+                        <svg class="w-3 h-3" aria-hidden="true" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 14 14">
+                            <path stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="m1 1 6 6m0 0 6 6M7 7l6-6M7 7l-6 6"/>
+                        </svg>
+                        <span class="sr-only">Close modal</span>
+                    </button>
+                </div>
+                <!-- Modal body -->
+                <div class="p-4 md:p-5 space-y-4">
+                    <div class="flex justify-center items-center">
+                    <img data-src="{{ $icon }}" alt="{{$model.Name}}" class="lazy rounded-t-lg max-h-48 max-w-96 object-cover mt-3">
+                  </div>
+
+                    <p class="text-base leading-relaxed text-gray-500 dark:text-gray-400">
+                    {{ $model.Description }}
+
+                    </p>
+
+                    <p class="text-base leading-relaxed text-gray-500 dark:text-gray-400">
+                    To install the model with the CLI, run: <br>
+                    <code> local-ai models install {{$model.Name}} </code> <br>
+
+                    <hr>
+                    See also <a href="https://localai.io/models/" target="_blank" >
+                    Installation <i class="fas fa-circle-info pr-2"></i>
+                    </a> to see how to install models with the REST API.
+                    </p>
+
+                    <p class="text-base leading-relaxed text-gray-500 dark:text-gray-400">
+                    <ul>
+                    {{ range $_, $u := $model.URLs }}
+                    <li><a href="{{ $u }}" target=_blank><i class="fa-solid fa-link"></i> {{ $u }}</a></li>
+                    {{ end }}
+                    </ul>
+                    </p>
+                </div>
+                <!-- Modal footer -->
+                <div class="flex items-center p-4 md:p-5 border-t border-gray-200 rounded-b dark:border-gray-600">
+                    <button data-modal-hide="{{ $model.Name}}-modal" type="button" class="py-2.5 px-5 ms-3 text-sm font-medium text-gray-900 focus:outline-none bg-white rounded-lg border border-gray-200 hover:bg-gray-100 hover:text-blue-700 focus:z-10 focus:ring-4 focus:ring-gray-100 dark:focus:ring-gray-700 dark:bg-gray-800 dark:text-gray-400 dark:border-gray-600 dark:hover:text-white dark:hover:bg-gray-700">Close</button>
+                </div>
+            </div>
+        </div>
+    </div>
+
+
+			</div>
+		</div>
+		</div>
+		{{ end }}
+
+		</div>
+  </div>
+</div>
+
+<script>
+var lazyLoadInstance = new LazyLoad({
+  // Your custom settings go here
+});
+
+let cards = document.querySelectorAll('.box')
+
+function liveSearch() {
+    let search_query = document.getElementById("searchbox").value;
+
+    //Use innerText if all contents are visible
+    //Use textContent for including hidden elements
+    for (var i = 0; i < cards.length; i++) {
+        if(cards[i].textContent.toLowerCase()
+                .includes(search_query.toLowerCase())) {
+            cards[i].classList.remove("is-hidden");
+        } else {
+            cards[i].classList.add("is-hidden");
+        }
+    }
+}
+
+//A little delay
+let typingTimer;
+let typeInterval = 500;
+let searchInput = document.getElementById('searchbox');
+
+searchInput.addEventListener('keyup', () => {
+    clearTimeout(typingTimer);
+    typingTimer = setTimeout(liveSearch, typeInterval);
+});
+</script>
+
+</div>
+
+<script src="https://cdnjs.cloudflare.com/ajax/libs/flowbite/2.3.0/flowbite.min.js"></script>
+</body>
+</html>
+`
+
+type GalleryModel struct {
+	Name        string   `json:"name" yaml:"name"`
+	URLs        []string `json:"urls" yaml:"urls"`
+	Icon        string   `json:"icon" yaml:"icon"`
+	Description string   `json:"description" yaml:"description"`
+}
+
+func main() {
+	// read the YAML file which contains the models
+
+	f, err := ioutil.ReadFile(os.Args[1])
+	if err != nil {
+		fmt.Println("Error reading file:", err)
+		return
+	}
+
+	models := []*GalleryModel{}
+	err = yaml.Unmarshal(f, &models)
+	if err != nil {
+		// write to stderr
+		os.Stderr.WriteString("Error unmarshaling YAML: " + err.Error() + "\n")
+		return
+	}
+
+	// Ensure that all arbitrary text content is sanitized before display
+	for i, m := range models {
+		models[i].Name = bluemonday.StrictPolicy().Sanitize(m.Name)
+		models[i].Description = bluemonday.StrictPolicy().Sanitize(m.Description)
+	}
+
+	// render the template
+	data := struct {
+		Models          []*GalleryModel
+		AvailableModels int
+	}{
+		Models:          models,
+		AvailableModels: len(models),
+	}
+	tmpl := template.Must(template.New("modelPage").Parse(modelPageTemplate))
+
+	err = tmpl.Execute(os.Stdout, data)
+	if err != nil {
+		fmt.Println("Error executing template:", err)
+		return
+	}
+}

.github/dependabot.yml

@@ -0,0 +1,123 @@
+# https://docs.github.com/en/code-security/dependabot/dependabot-version-updates/configuration-options-for-the-dependabot.yml-file
+version: 2
+updates:
+  - package-ecosystem: "gitsubmodule"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "gomod"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+    ignore:
+    - dependency-name: "github.com/mudler/LocalAI/pkg/grpc/proto"
+  - package-ecosystem: "github-actions"
+    # Workflow files stored in the default location of `.github/workflows`. (You don't need to specify `/.github/workflows` for `directory`. You can use `directory: "/"`.)
+    directory: "/"
+    schedule:
+      # Check for updates to GitHub Actions every weekday
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    # Workflow files stored in the default location of `.github/workflows`. (You don't need to specify `/.github/workflows` for `directory`. You can use `directory: "/"`.)
+    directory: "/"
+    schedule:
+      # Check for updates to GitHub Actions every weekday
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    # Workflow files stored in the default location of `.github/workflows`. (You don't need to specify `/.github/workflows` for `directory`. You can use `directory: "/"`.)
+    directory: "/"
+    schedule:
+      # Check for updates to GitHub Actions every weekday
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/bark"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/common/template"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/coqui"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/diffusers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/exllama"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/exllama2"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/mamba"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/openvoice"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/parler-tts"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/rerankers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/sentencetransformers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/transformers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/vllm"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/chainlit"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/functions"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/langchain/langchainpy-localai-example"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/langchain-chroma"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/streamlit-bot"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/k8sgpt"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/kubernetes"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/langchain"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "gomod"
+    directory: "/examples/semantic-todo"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/telegram-bot"
+    schedule:
+      interval: "weekly"

.github/labeler.yml

@@ -0,0 +1,33 @@
+enhancement:
+ - head-branch: ['^feature', 'feature']
+
+dependencies:
+- any:
+  - changed-files:
+    - any-glob-to-any-file: 'Makefile'
+  - changed-files:
+    - any-glob-to-any-file: '*.mod'
+  - changed-files:
+    - any-glob-to-any-file: '*.sum'
+
+kind/documentation:
+- any:
+  - changed-files:
+    - any-glob-to-any-file: 'docs/*'
+  - changed-files:
+    - any-glob-to-any-file: '*.md'
+
+area/ai-model:
+- any:
+  - changed-files:
+    - any-glob-to-any-file: 'gallery/*'
+
+examples:
+- any:
+  - changed-files:
+    - any-glob-to-any-file: 'examples/*'
+
+ci:
+- any:
+  - changed-files:
+    - any-glob-to-any-file: '.github/*'

.github/release.yml

@@ -0,0 +1,37 @@
+# .github/release.yml
+
+changelog:
+  exclude:
+    labels:
+      - ignore-for-release
+  categories:
+    - title: Breaking Changes 🛠
+      labels:
+        - Semver-Major
+        - breaking-change
+    - title: "Bug fixes :bug:"
+      labels:
+        - bug
+        - regression
+    - title: "🖧 P2P area"
+      labels:
+         - area/p2p
+    - title: Exciting New Features 🎉
+      labels:
+        - Semver-Minor
+        - enhancement
+        - ux
+        - roadmap
+    - title: 🧠 Models
+      labels:
+        - area/ai-model
+    - title: 📖 Documentation and examples
+      labels:
+        - kind/documentation
+        - examples
+    - title: 👒 Dependencies
+      labels:
+        - dependencies
+    - title: Other Changes
+      labels:
+        - "*"

.github/stale.yml

@@ -0,0 +1,18 @@
+# Number of days of inactivity before an issue becomes stale
+daysUntilStale: 45
+# Number of days of inactivity before a stale issue is closed
+daysUntilClose: 10
+# Issues with these labels will never be considered stale
+exemptLabels:
+  - issue/willfix
+# Label to use when marking an issue as stale
+staleLabel: issue/stale
+# Comment to post when marking an issue as stale. Set to `false` to disable
+markComment: >
+  This issue has been automatically marked as stale because it has not had
+  recent activity. It will be closed if no further activity occurs. Thank you
+  for your contributions.
+# Comment to post when closing a stale issue. Set to `false` to disable
+closeComment: >
+  This issue is being automatically closed due to inactivity.
+  However, you may choose to reopen this issue.

.github/workflows/bump_deps.yaml

@@ -9,42 +9,52 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - repository: "go-skynet/go-llama.cpp"
-            variable: "GOLLAMA_VERSION"
+          - repository: "ggml-org/llama.cpp"
+            variable: "CPPLLAMA_VERSION"
             branch: "master"
-          - repository: "go-skynet/go-gpt2.cpp"
-            variable: "GOGPT2_VERSION"
-            branch: "master"
-          - repository: "donomii/go-rwkv.cpp"
-            variable: "RWKV_VERSION"
-            branch: "main"
           - repository: "ggerganov/whisper.cpp"
             variable: "WHISPER_CPP_VERSION"
             branch: "master"
-          - repository: "go-skynet/go-bert.cpp"
-            variable: "BERT_VERSION"
+          - repository: "PABannier/bark.cpp"
+            variable: "BARKCPP_VERSION"
+            branch: "main"
+          - repository: "leejet/stable-diffusion.cpp"
+            variable: "STABLEDIFFUSION_GGML_VERSION"
+            branch: "master"
+          - repository: "mudler/go-stable-diffusion"
+            variable: "STABLEDIFFUSION_VERSION"
+            branch: "master"
+          - repository: "mudler/go-piper"
+            variable: "PIPER_VERSION"
             branch: "master"
-          - repository: "go-skynet/bloomz.cpp"
-            variable: "BLOOMZ_VERSION"
-            branch: "main"
-          - repository: "nomic-ai/gpt4all"
-            variable: "GPT4ALL_VERSION"
-            branch: "main"
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Bump dependencies 🔧
+        id: bump
         run: |
           bash .github/bump_deps.sh ${{ matrix.repository }} ${{ matrix.branch }} ${{ matrix.variable }}
+          {
+            echo 'message<<EOF'
+            cat "${{ matrix.variable }}_message.txt"
+            echo EOF
+          } >> "$GITHUB_OUTPUT"
+          {
+            echo 'commit<<EOF'
+            cat "${{ matrix.variable }}_commit.txt"
+            echo EOF
+          } >> "$GITHUB_OUTPUT"
+          rm -rfv ${{ matrix.variable }}_message.txt
+          rm -rfv ${{ matrix.variable }}_commit.txt
       - name: Create Pull Request
-        uses: peter-evans/create-pull-request@v5
+        uses: peter-evans/create-pull-request@v7
         with:
           token: ${{ secrets.UPDATE_BOT_TOKEN }}
           push-to-fork: ci-forks/LocalAI
           commit-message: ':arrow_up: Update ${{ matrix.repository }}'
-          title: ':arrow_up: Update ${{ matrix.repository }}'
+          title: 'chore: :arrow_up: Update ${{ matrix.repository }} to `${{ steps.bump.outputs.commit }}`'
           branch: "update/${{ matrix.variable }}"
-          body: Bump of ${{ matrix.repository }} version
+          body: ${{ steps.bump.outputs.message }}
           signoff: true
 
 

.github/workflows/bump_docs.yaml

@@ -0,0 +1,31 @@
+name: Bump dependencies
+on:
+  schedule:
+    - cron: 0 20 * * *
+  workflow_dispatch:
+jobs:
+  bump:
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - repository: "mudler/LocalAI"
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Bump dependencies 🔧
+        run: |
+          bash .github/bump_docs.sh ${{ matrix.repository }}
+      - name: Create Pull Request
+        uses: peter-evans/create-pull-request@v7
+        with:
+          token: ${{ secrets.UPDATE_BOT_TOKEN }}
+          push-to-fork: ci-forks/LocalAI
+          commit-message: ':arrow_up: Update docs version ${{ matrix.repository }}'
+          title: 'docs: :arrow_up: update docs version ${{ matrix.repository }}'
+          branch: "update/docs"
+          body: Bump of ${{ matrix.repository }} version inside docs
+          signoff: true
+
+
+

.github/workflows/checksum_checker.yaml

@@ -0,0 +1,47 @@
+name: Check if checksums are up-to-date
+on:
+  schedule:
+    - cron: 0 20 * * *
+  workflow_dispatch:
+jobs:
+  checksum_check:
+    runs-on: arc-runner-set
+    steps:
+      - name: Force Install GIT latest
+        run: |
+          sudo apt-get update \
+          && sudo apt-get install -y software-properties-common \
+          && sudo apt-get update \
+          && sudo add-apt-repository -y ppa:git-core/ppa \
+          && sudo apt-get update \
+          && sudo apt-get install -y git
+      - uses: actions/checkout@v4
+      - name: Install dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y pip wget
+          sudo pip install --upgrade pip
+          pip install huggingface_hub
+      - name: 'Setup yq'
+        uses: dcarbone/install-yq-action@v1.3.1
+        with:
+          version: 'v4.44.2'
+          download-compressed: true
+          force: true
+
+      - name: Checksum checker 🔧
+        run: |
+          export HF_HOME=/hf_cache
+          sudo mkdir /hf_cache
+          sudo chmod 777 /hf_cache
+          bash .github/checksum_checker.sh gallery/index.yaml
+      - name: Create Pull Request
+        uses: peter-evans/create-pull-request@v7
+        with:
+          token: ${{ secrets.UPDATE_BOT_TOKEN }}
+          push-to-fork: ci-forks/LocalAI
+          commit-message: ':arrow_up: Checksum updates in gallery/index.yaml'
+          title: 'chore(model-gallery): :arrow_up: update checksum'
+          branch: "update/checksum"
+          body: Updating checksums in gallery/index.yaml
+          signoff: true

.github/workflows/dependabot_auto.yml

@@ -0,0 +1,43 @@
+name: Dependabot auto-merge
+on:
+- pull_request_target
+
+permissions:
+  contents: write
+  pull-requests: write
+  packages: read
+
+jobs:
+  dependabot:
+    runs-on: ubuntu-latest
+    if: ${{ github.actor == 'dependabot[bot]' }}
+    steps:
+      - name: Dependabot metadata
+        id: metadata
+        uses: dependabot/fetch-metadata@v2.3.0
+        with:
+          github-token: "${{ secrets.GITHUB_TOKEN }}"
+          skip-commit-verification: true
+
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Approve a PR if not already approved
+        run: |
+          gh pr checkout "$PR_URL"
+            if [ "$(gh pr status --json reviewDecision -q .currentBranch.reviewDecision)" != "APPROVED" ];
+          then
+            gh pr review --approve "$PR_URL"
+          else
+            echo "PR already approved.";
+          fi
+        env:
+          PR_URL: ${{github.event.pull_request.html_url}}
+          GITHUB_TOKEN: ${{secrets.GITHUB_TOKEN}}
+
+      - name: Enable auto-merge for Dependabot PRs
+        if: ${{ contains(github.event.pull_request.title, 'bump')}}
+        run: gh pr merge --auto --squash "$PR_URL"
+        env:
+          PR_URL: ${{github.event.pull_request.html_url}}
+          GITHUB_TOKEN: ${{secrets.GITHUB_TOKEN}}

.github/workflows/deploy-explorer.yaml

@@ -0,0 +1,64 @@
+name: Explorer deployment
+
+on:
+  push:
+    branches:
+      - master
+    tags:
+      - 'v*'
+
+concurrency:
+  group: ci-deploy-${{ github.head_ref || github.ref }}-${{ github.repository }}
+
+jobs:
+  build-linux:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - uses: actions/setup-go@v5
+        with:
+          go-version: '1.21.x'
+          cache: false
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y wget curl build-essential ffmpeg protobuf-compiler ccache upx-ucl gawk cmake libgmock-dev
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          make protogen-go
+      - name: Build api
+        run: |
+          CGO_ENABLED=0 make build-api
+      - name: rm
+        uses: appleboy/ssh-action@v1.2.2
+        with:
+            host: ${{ secrets.EXPLORER_SSH_HOST }}
+            username: ${{ secrets.EXPLORER_SSH_USERNAME }}
+            key: ${{ secrets.EXPLORER_SSH_KEY }}
+            port: ${{ secrets.EXPLORER_SSH_PORT }}
+            script: |
+                sudo rm -rf local-ai/ || true
+      - name: copy file via ssh
+        uses: appleboy/scp-action@v0.1.7
+        with:
+            host: ${{ secrets.EXPLORER_SSH_HOST }}
+            username: ${{ secrets.EXPLORER_SSH_USERNAME }}
+            key: ${{ secrets.EXPLORER_SSH_KEY }}
+            port: ${{ secrets.EXPLORER_SSH_PORT }}
+            source: "local-ai"
+            overwrite: true
+            rm: true
+            target: ./local-ai
+      - name: restarting
+        uses: appleboy/ssh-action@v1.2.2
+        with:
+            host: ${{ secrets.EXPLORER_SSH_HOST }}
+            username: ${{ secrets.EXPLORER_SSH_USERNAME }}
+            key: ${{ secrets.EXPLORER_SSH_KEY }}
+            port: ${{ secrets.EXPLORER_SSH_PORT }}
+            script: |
+                sudo cp -rfv local-ai/local-ai /usr/bin/local-ai
+                sudo systemctl restart local-ai

.github/workflows/disabled/comment-pr.yaml

@@ -0,0 +1,83 @@
+name: Comment PRs
+on:
+  pull_request_target:
+
+jobs:
+  comment-pr:
+    env:
+        MODEL_NAME: hermes-2-theta-llama-3-8b
+    runs-on: ubuntu-latest
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v3
+      with:
+        ref: "${{ github.event.pull_request.merge_commit_sha }}"
+        fetch-depth: 0 # needed to checkout all branches for this Action to work
+    - uses: mudler/localai-github-action@v1
+      with:
+        model: 'hermes-2-theta-llama-3-8b' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
+      # Check the PR diff using the current branch and the base branch of the PR
+    - uses: GrantBirki/git-diff-action@v2.7.0
+      id: git-diff-action
+      with:
+            json_diff_file_output: diff.json
+            raw_diff_file_output: diff.txt
+            file_output_only: "true"
+            base_branch: ${{ github.event.pull_request.base.sha }}
+    - name: Show diff
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      run: |
+            cat $DIFF
+    - name: Summarize
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      id: summarize
+      run: |
+            input="$(cat $DIFF)"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "You are LocalAI-bot in Github that helps understanding PRs and assess complexity. Explain what has changed in this PR diff and why"
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary="$(echo $response | jq -r '.choices[0].message.content')"
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            echo "Summary:"
+            echo "$summary"
+            echo "payload sent"
+            echo "$json_payload"
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+            docker logs --tail 10 local-ai
+    - uses: mshick/add-pr-comment@v2
+      if: always()
+      with:
+          repo-token: ${{ secrets.UPDATE_BOT_TOKEN }}
+          message: ${{ steps.summarize.outputs.message }}
+          message-failure: |
+            Uh oh! Could not analyze this PR, maybe it's too big?

.github/workflows/disabled/test-gpu.yml

@@ -0,0 +1,63 @@
+---
+name: 'GPU tests'
+
+on:
+  pull_request:
+  push:
+    branches:
+      - master
+    tags:
+      - '*'
+
+concurrency:
+  group: ci-gpu-tests-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
+jobs:
+  ubuntu-latest:
+    runs-on: gpu
+    strategy:
+      matrix:
+        go-version: ['1.21.x']
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with: 
+          submodules: true
+      - name: Setup Go ${{ matrix.go-version }}
+        uses: actions/setup-go@v4
+        with:
+          go-version: ${{ matrix.go-version }}
+      # You can test your matrix by printing the current Go version
+      - name: Display Go version
+        run: go version
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo DEBIAN_FRONTEND=noninteractive apt-get install -y make wget
+      - name: Build
+        run: |
+          if [ ! -e /run/systemd/system ]; then
+            sudo mkdir /run/systemd/system
+          fi
+          sudo mkdir -p /host/tests/${{ github.head_ref || github.ref }}
+          sudo chmod -R 777 /host/tests/${{ github.head_ref || github.ref }}
+          make \
+            TEST_DIR="/host/tests/${{ github.head_ref || github.ref }}" \
+            BUILD_TYPE=cublas \
+            prepare-e2e run-e2e-image test-e2e
+      - name: Release space from worker ♻
+        if: always()
+        run: |
+          sudo rm -rf build || true
+          sudo rm -rf bin || true
+          sudo rm -rf dist || true
+          sudo docker logs $(sudo docker ps -q --filter ancestor=localai-tests) > logs.txt
+          sudo cat logs.txt || true
+          sudo rm -rf logs.txt
+          make clean || true
+          make \
+            TEST_DIR="/host/tests/${{ github.head_ref || github.ref }}" \
+            teardown-e2e || true
+          sudo rm -rf /host/tests/${{ github.head_ref || github.ref }} || true
+          docker system prune -f -a --volumes || true

.github/workflows/generate_grpc_cache.yaml

@@ -0,0 +1,95 @@
+name: 'generate and publish GRPC docker caches'
+
+on:
+  workflow_dispatch:
+
+  schedule:
+    # daily at midnight
+    - cron: '0 0 * * *'
+
+concurrency:
+  group: grpc-cache-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
+jobs:
+  generate_caches:
+    strategy:
+      matrix:
+        include:
+          - grpc-base-image: ubuntu:22.04
+            runs-on: 'arc-runner-set'
+            platforms: 'linux/amd64,linux/arm64'
+    runs-on: ${{matrix.runs-on}}
+    steps:
+      - name: Release space from worker
+        if: matrix.runs-on == 'ubuntu-latest'
+        run: |
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          df -h
+          echo
+          sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
+          sudo apt-get remove --auto-remove android-sdk-platform-tools || true
+          sudo apt-get purge --auto-remove android-sdk-platform-tools || true
+          sudo rm -rf /usr/local/lib/android
+          sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
+          sudo rm -rf /usr/share/dotnet
+          sudo apt-get remove -y '^mono-.*' || true
+          sudo apt-get remove -y '^ghc-.*' || true
+          sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
+          sudo apt-get remove -y 'php.*' || true
+          sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
+          sudo apt-get remove -y '^google-.*' || true
+          sudo apt-get remove -y azure-cli || true
+          sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
+          sudo apt-get remove -y '^gfortran-.*' || true
+          sudo apt-get remove -y microsoft-edge-stable || true
+          sudo apt-get remove -y firefox || true
+          sudo apt-get remove -y powershell || true
+          sudo apt-get remove -y r-base-core || true
+          sudo apt-get autoremove -y
+          sudo apt-get clean
+          echo
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          sudo rm -rfv build || true
+          sudo rm -rf /usr/share/dotnet || true
+          sudo rm -rf /opt/ghc || true
+          sudo rm -rf "/usr/local/share/boost" || true
+          sudo rm -rf "$AGENT_TOOLSDIRECTORY" || true
+          df -h
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@master
+        with:
+          platforms: all
+
+      - name: Set up Docker Buildx
+        id: buildx
+        uses: docker/setup-buildx-action@master
+
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Cache GRPC
+        uses: docker/build-push-action@v6
+        with:
+          builder: ${{ steps.buildx.outputs.name }}
+          # The build-args MUST be an EXACT match between the image cache and other workflow steps that want to use that cache.
+          # This means that even the MAKEFLAGS have to be an EXACT match.
+          # If the build-args are not an EXACT match, it will result in a cache miss, which will require GRPC to be built from scratch.
+          build-args: |
+            GRPC_BASE_IMAGE=${{ matrix.grpc-base-image }}
+            GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
+            GRPC_VERSION=v1.65.0
+          context: .
+          file: ./Dockerfile
+          cache-to: type=gha,ignore-error=true
+          cache-from: type=gha
+          target: grpc
+          platforms: ${{ matrix.platforms }}
+          push: false

.github/workflows/generate_intel_image.yaml

@@ -0,0 +1,59 @@
+name: 'generate and publish intel docker caches'
+
+on:
+  workflow_dispatch:
+  push:
+    branches:
+      - master
+
+concurrency:
+  group: intel-cache-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
+jobs:
+  generate_caches:
+    strategy:
+      matrix:
+        include:
+          - base-image: intel/oneapi-basekit:2025.1.0-0-devel-ubuntu22.04
+            runs-on: 'ubuntu-latest'
+            platforms: 'linux/amd64'
+    runs-on: ${{matrix.runs-on}}
+    steps:
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@master
+        with:
+          platforms: all
+      - name: Login to DockerHub
+        if: github.event_name != 'pull_request'
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_PASSWORD }}
+
+      - name: Login to quay
+        if: github.event_name != 'pull_request'
+        uses: docker/login-action@v3
+        with:
+          registry: quay.io
+          username: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+          password: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+      - name: Set up Docker Buildx
+        id: buildx
+        uses: docker/setup-buildx-action@master
+
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Cache Intel images
+        uses: docker/build-push-action@v6
+        with:
+          builder: ${{ steps.buildx.outputs.name }}
+          build-args: |
+            BASE_IMAGE=${{ matrix.base-image }}
+          context: .
+          file: ./Dockerfile
+          tags: quay.io/go-skynet/intel-oneapi-base:latest
+          push: true
+          target: intel
+          platforms: ${{ matrix.platforms }}

.github/workflows/image-pr.yml

@@ -0,0 +1,140 @@
+---
+name: 'build container images tests'
+
+on:
+  pull_request:
+
+concurrency:
+  group: ci-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
+jobs:
+  extras-image-build:
+    uses: ./.github/workflows/image_build.yml
+    with:
+      tag-latest: ${{ matrix.tag-latest }}
+      tag-suffix: ${{ matrix.tag-suffix }}
+      ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
+      build-type: ${{ matrix.build-type }}
+      cuda-major-version: ${{ matrix.cuda-major-version }}
+      cuda-minor-version: ${{ matrix.cuda-minor-version }}
+      platforms: ${{ matrix.platforms }}
+      runs-on: ${{ matrix.runs-on }}
+      base-image: ${{ matrix.base-image }}
+      grpc-base-image: ${{ matrix.grpc-base-image }}
+      makeflags: ${{ matrix.makeflags }}
+    secrets:
+      dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
+      dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
+      quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+      quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+    strategy:
+      # Pushing with all jobs in parallel
+      # eats the bandwidth of all the nodes
+      max-parallel: ${{ github.event_name != 'pull_request' && 4 || 8 }}
+      matrix:
+        include:
+          # This is basically covered by the AIO test
+          # - build-type: ''
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-ffmpeg'
+          #   ffmpeg: 'true'
+          #   image-type: 'extras'
+          #   runs-on: 'arc-runner-set'
+          #   base-image: "ubuntu:22.04"
+          #   makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda12-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          # - build-type: 'hipblas'
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-hipblas'
+          #   ffmpeg: 'false'
+          #   image-type: 'extras'
+          #   base-image: "rocm/dev-ubuntu-22.04:6.1"
+          #   grpc-base-image: "ubuntu:22.04"
+          #   runs-on: 'arc-runner-set'
+          #   makeflags: "--jobs=3 --output-sync=target"
+          # - build-type: 'sycl_f16'
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+          #   grpc-base-image: "ubuntu:22.04"
+          #   tag-suffix: 'sycl-f16-ffmpeg'
+          #   ffmpeg: 'true'
+          #   image-type: 'extras'
+          #   runs-on: 'arc-runner-set'
+          #   makeflags: "--jobs=3 --output-sync=target"
+  # core-image-build:
+  #   uses: ./.github/workflows/image_build.yml
+  #   with:
+  #     tag-latest: ${{ matrix.tag-latest }}
+  #     tag-suffix: ${{ matrix.tag-suffix }}
+  #     ffmpeg: ${{ matrix.ffmpeg }}
+  #     image-type: ${{ matrix.image-type }}
+  #     build-type: ${{ matrix.build-type }}
+  #     cuda-major-version: ${{ matrix.cuda-major-version }}
+  #     cuda-minor-version: ${{ matrix.cuda-minor-version }}
+  #     platforms: ${{ matrix.platforms }}
+  #     runs-on: ${{ matrix.runs-on }}
+  #     base-image: ${{ matrix.base-image }}
+  #     grpc-base-image: ${{ matrix.grpc-base-image }}
+  #     makeflags: ${{ matrix.makeflags }}
+  #   secrets:
+  #     dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
+  #     dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
+  #     quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+  #     quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+  #   strategy:
+  #     matrix:
+  #       include:
+          # - build-type: ''
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-ffmpeg-core'
+          #   ffmpeg: 'true'
+          #   image-type: 'core'
+          #   runs-on: 'ubuntu-latest'
+          #   base-image: "ubuntu:22.04"
+          #   makeflags: "--jobs=4 --output-sync=target"
+          # - build-type: 'sycl_f16'
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+          #   grpc-base-image: "ubuntu:22.04"
+          #   tag-suffix: 'sycl-f16-ffmpeg-core'
+          #   ffmpeg: 'true'
+          #   image-type: 'core'
+          #   runs-on: 'arc-runner-set'
+          #   makeflags: "--jobs=3 --output-sync=target"
+          # - build-type: 'cublas'
+          #   cuda-major-version: "12"
+          #   cuda-minor-version: "0"
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-cublas-cuda12-ffmpeg-core'
+          #   ffmpeg: 'true'
+          #   image-type: 'core'
+          #   runs-on: 'ubuntu-latest'
+          #   base-image: "ubuntu:22.04"
+          #   makeflags: "--jobs=4 --output-sync=target"
+          # - build-type: 'vulkan'
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-vulkan-ffmpeg-core'
+          #   ffmpeg: 'true'
+          #   image-type: 'core'
+          #   runs-on: 'ubuntu-latest'
+          #   base-image: "ubuntu:22.04"
+          #   makeflags: "--jobs=4 --output-sync=target"

.github/workflows/image.yml

@@ -2,77 +2,408 @@
 name: 'build container images'
 
 on:
-  pull_request:
   push:
     branches:
       - master
     tags:
       - '*'
 
+concurrency:
+  group: ci-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
 jobs:
-  docker:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
+  hipblas-jobs:
+    uses: ./.github/workflows/image_build.yml
+    with:
+      tag-latest: ${{ matrix.tag-latest }}
+      tag-suffix: ${{ matrix.tag-suffix }}
+      ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
+      build-type: ${{ matrix.build-type }}
+      cuda-major-version: ${{ matrix.cuda-major-version }}
+      cuda-minor-version: ${{ matrix.cuda-minor-version }}
+      platforms: ${{ matrix.platforms }}
+      runs-on: ${{ matrix.runs-on }}
+      base-image: ${{ matrix.base-image }}
+      grpc-base-image: ${{ matrix.grpc-base-image }}
+      aio: ${{ matrix.aio }}
+      makeflags: ${{ matrix.makeflags }}
+      latest-image: ${{ matrix.latest-image }}
+      latest-image-aio: ${{ matrix.latest-image-aio }}
+    secrets:
+      dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
+      dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
+      quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+      quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+    strategy:
+      # Pushing with all jobs in parallel
+      # eats the bandwidth of all the nodes
+      max-parallel: 2
+      matrix:
+        include:
+          - build-type: 'hipblas'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-hipblas-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            aio: "-aio-gpu-hipblas"
+            base-image: "rocm/dev-ubuntu-22.04:6.1"
+            grpc-base-image: "ubuntu:22.04"
+            latest-image: 'latest-gpu-hipblas'
+            latest-image-aio: 'latest-aio-gpu-hipblas'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'hipblas'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-hipblas'
+            ffmpeg: 'false'
+            image-type: 'extras'
+            base-image: "rocm/dev-ubuntu-22.04:6.1"
+            grpc-base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'hipblas'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-hipblas-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            base-image: "rocm/dev-ubuntu-22.04:6.1"
+            grpc-base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+            latest-image: 'latest-gpu-hipblas-core'
+          - build-type: 'hipblas'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-hipblas-core'
+            ffmpeg: 'false'
+            image-type: 'core'
+            base-image: "rocm/dev-ubuntu-22.04:6.1"
+            grpc-base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+  self-hosted-jobs:
+    uses: ./.github/workflows/image_build.yml
+    with:
+      tag-latest: ${{ matrix.tag-latest }}
+      tag-suffix: ${{ matrix.tag-suffix }}
+      ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
+      build-type: ${{ matrix.build-type }}
+      cuda-major-version: ${{ matrix.cuda-major-version }}
+      cuda-minor-version: ${{ matrix.cuda-minor-version }}
+      platforms: ${{ matrix.platforms }}
+      runs-on: ${{ matrix.runs-on }}
+      base-image: ${{ matrix.base-image }}
+      grpc-base-image: ${{ matrix.grpc-base-image }}
+      aio: ${{ matrix.aio }}
+      makeflags: ${{ matrix.makeflags }}
+      latest-image: ${{ matrix.latest-image }}
+      latest-image-aio: ${{ matrix.latest-image-aio }}
+    secrets:
+      dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
+      dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
+      quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+      quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+    strategy:
+      # Pushing with all jobs in parallel
+      # eats the bandwidth of all the nodes
+      max-parallel: ${{ github.event_name != 'pull_request' && 5 || 8 }}
+      matrix:
+        include:
+          # Extra images
+          - build-type: ''
+            #platforms: 'linux/amd64,linux/arm64'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: ''
+            ffmpeg: ''
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: ''
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda11'
+            ffmpeg: ''
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda12'
+            ffmpeg: ''
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-cublas-cuda11-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            aio: "-aio-gpu-nvidia-cuda-11"
+            latest-image: 'latest-gpu-nvidia-cuda-11'
+            latest-image-aio: 'latest-aio-gpu-nvidia-cuda-11'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-cublas-cuda12-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            aio: "-aio-gpu-nvidia-cuda-12"
+            latest-image: 'latest-gpu-nvidia-cuda-12'
+            latest-image-aio: 'latest-aio-gpu-nvidia-cuda-12'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: ''
+            #platforms: 'linux/amd64,linux/arm64'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: ''
+            ffmpeg: ''
+            image-type: 'extras'
+            base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f16'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f16-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            aio: "-aio-gpu-intel-f16"
+            latest-image: 'latest-gpu-intel-f16'
+            latest-image-aio: 'latest-aio-gpu-intel-f16'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f32'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f32-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            aio: "-aio-gpu-intel-f32"
+            latest-image: 'latest-gpu-intel-f32'
+            latest-image-aio: 'latest-aio-gpu-intel-f32'
+            makeflags: "--jobs=3 --output-sync=target"
+          # Core images
+          - build-type: 'sycl_f16'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f16-core'
+            ffmpeg: 'false'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f32'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f32-core'
+            ffmpeg: 'false'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f16'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f16-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+            latest-image: 'latest-gpu-intel-f16-core'
+          - build-type: 'sycl_f32'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f32-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+            latest-image: 'latest-gpu-intel-f32-core'
 
-      - name: Prepare
-        id: prep
-        run: |
-          DOCKER_IMAGE=quay.io/go-skynet/local-ai
-          VERSION=master
-          SHORTREF=${GITHUB_SHA::8}
-
-          # If this is git tag, use the tag name as a docker tag
-          if [[ $GITHUB_REF == refs/tags/* ]]; then
-            VERSION=${GITHUB_REF#refs/tags/}
-          fi
-          TAGS="${DOCKER_IMAGE}:${VERSION},${DOCKER_IMAGE}:${SHORTREF}"
-
-          # If the VERSION looks like a version number, assume that
-          # this is the most recent version of the image and also
-          # tag it 'latest'.
-          if [[ $VERSION =~ ^v[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}$ ]]; then
-            TAGS="$TAGS,${DOCKER_IMAGE}:latest"
-          fi
-
-          # Set output parameters.
-          echo ::set-output name=tags::${TAGS}
-          echo ::set-output name=docker_image::${DOCKER_IMAGE}
-
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@master
-        with:
-          platforms: all
-
-      - name: Set up Docker Buildx
-        id: buildx
-        uses: docker/setup-buildx-action@master
-
-      - name: Login to DockerHub
-        if: github.event_name != 'pull_request'
-        uses: docker/login-action@v2
-        with:
-          registry: quay.io
-          username: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
-          password: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
-      - name: Build
-        if: github.event_name != 'pull_request'
-        uses: docker/build-push-action@v4
-        with:
-          builder: ${{ steps.buildx.outputs.name }}
-          context: .
-          file: ./Dockerfile
-          platforms: linux/amd64,linux/arm64
-          push: true
-          tags: ${{ steps.prep.outputs.tags }}
-      - name: Build PRs
-        if: github.event_name == 'pull_request'
-        uses: docker/build-push-action@v4
-        with:
-          builder: ${{ steps.buildx.outputs.name }}
-          context: .
-          file: ./Dockerfile
-          platforms: linux/amd64
-          push: false
-          tags: ${{ steps.prep.outputs.tags }}
+  core-image-build:
+    uses: ./.github/workflows/image_build.yml
+    with:
+      tag-latest: ${{ matrix.tag-latest }}
+      tag-suffix: ${{ matrix.tag-suffix }}
+      ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
+      build-type: ${{ matrix.build-type }}
+      cuda-major-version: ${{ matrix.cuda-major-version }}
+      cuda-minor-version: ${{ matrix.cuda-minor-version }}
+      platforms: ${{ matrix.platforms }}
+      runs-on: ${{ matrix.runs-on }}
+      aio: ${{ matrix.aio }}
+      base-image: ${{ matrix.base-image }}
+      grpc-base-image: ${{ matrix.grpc-base-image }}
+      makeflags: ${{ matrix.makeflags }}
+      latest-image: ${{ matrix.latest-image }}
+      latest-image-aio: ${{ matrix.latest-image-aio }}
+      skip-drivers: ${{ matrix.skip-drivers }}
+    secrets:
+      dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
+      dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
+      quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+      quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+    strategy:
+      max-parallel: ${{ github.event_name != 'pull_request' && 2 || 4 }}
+      matrix:
+        include:
+          - build-type: ''
+            platforms: 'linux/amd64,linux/arm64'
+            tag-latest: 'auto'
+            tag-suffix: '-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            aio: "-aio-cpu"
+            latest-image: 'latest-cpu'
+            latest-image-aio: 'latest-aio-cpu'
+            makeflags: "--jobs=4 --output-sync=target"
+            skip-drivers: 'false'
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda11-core'
+            ffmpeg: ''
+            image-type: 'core'
+            base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=4 --output-sync=target"
+            skip-drivers: 'false'
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda12-core'
+            ffmpeg: ''
+            image-type: 'core'
+            base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=4 --output-sync=target"
+            skip-drivers: 'false'
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda11-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=4 --output-sync=target"
+            skip-drivers: 'false'
+            latest-image: 'latest-gpu-nvidia-cuda-12-core'
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda12-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            skip-drivers: 'false'
+            makeflags: "--jobs=4 --output-sync=target"
+            latest-image: 'latest-gpu-nvidia-cuda-12-core'
+          - build-type: 'vulkan'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-vulkan-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            skip-drivers: 'false'
+            makeflags: "--jobs=4 --output-sync=target"
+            latest-image: 'latest-gpu-vulkan-core'
+  gh-runner:
+    uses: ./.github/workflows/image_build.yml
+    with:
+      tag-latest: ${{ matrix.tag-latest }}
+      tag-suffix: ${{ matrix.tag-suffix }}
+      ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
+      build-type: ${{ matrix.build-type }}
+      cuda-major-version: ${{ matrix.cuda-major-version }}
+      cuda-minor-version: ${{ matrix.cuda-minor-version }}
+      platforms: ${{ matrix.platforms }}
+      runs-on: ${{ matrix.runs-on }}
+      aio: ${{ matrix.aio }}
+      base-image: ${{ matrix.base-image }}
+      grpc-base-image: ${{ matrix.grpc-base-image }}
+      makeflags: ${{ matrix.makeflags }}
+      latest-image: ${{ matrix.latest-image }}
+      latest-image-aio: ${{ matrix.latest-image-aio }}
+      skip-drivers: ${{ matrix.skip-drivers }}
+    secrets:
+      dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
+      dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
+      quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+      quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+    strategy:
+      matrix:
+        include:
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/arm64'
+            tag-latest: 'false'
+            tag-suffix: '-nvidia-l4t-arm64-core'
+            latest-image: 'latest-nvidia-l4t-arm64-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            base-image: "nvcr.io/nvidia/l4t-jetpack:r36.4.0"
+            runs-on: 'ubuntu-24.04-arm'
+            makeflags: "--jobs=4 --output-sync=target"
+            skip-drivers: 'true'

.github/workflows/image_build.yml

@@ -0,0 +1,346 @@
+---
+name: 'build container images (reusable)'
+
+on:
+  workflow_call:
+    inputs:
+      base-image:
+        description: 'Base image'
+        required: true
+        type: string
+      grpc-base-image:
+        description: 'GRPC Base image, must be a compatible image with base-image'
+        required: false
+        default: ''
+        type: string
+      build-type:
+        description: 'Build type'
+        default: ''
+        type: string
+      cuda-major-version:
+        description: 'CUDA major version'
+        default: "12"
+        type: string
+      cuda-minor-version:
+        description: 'CUDA minor version'
+        default: "4"
+        type: string
+      platforms:
+        description: 'Platforms'
+        default: ''
+        type: string
+      tag-latest:
+        description: 'Tag latest'
+        default: ''
+        type: string
+      latest-image:
+          description: 'Tag latest'
+          default: ''
+          type: string
+      latest-image-aio:
+          description: 'Tag latest'
+          default: ''
+          type: string
+      tag-suffix:
+        description: 'Tag suffix'
+        default: ''
+        type: string
+      ffmpeg:
+        description: 'FFMPEG'
+        default: ''
+        type: string
+      skip-drivers:
+        description: 'Skip drivers by default'
+        default: 'false'
+        type: string
+      image-type:
+        description: 'Image type'
+        default: ''
+        type: string
+      runs-on:
+        description: 'Runs on'
+        required: true
+        default: ''
+        type: string
+      makeflags:
+        description: 'Make Flags'
+        required: false
+        default: '--jobs=4 --output-sync=target'
+        type: string
+      aio:
+        description: 'AIO Image Name'
+        required: false
+        default: ''
+        type: string
+    secrets:
+      dockerUsername:
+        required: true
+      dockerPassword:
+        required: true
+      quayUsername:
+        required: true
+      quayPassword:
+        required: true
+jobs:
+  reusable_image-build:
+    runs-on: ${{ inputs.runs-on }}
+    steps:
+      - name: Force Install GIT latest
+        run: |
+          sudo apt-get update \
+          && sudo apt-get install -y software-properties-common \
+          && sudo apt-get update \
+          && sudo add-apt-repository -y ppa:git-core/ppa \
+          && sudo apt-get update \
+          && sudo apt-get install -y git
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Release space from worker
+        if: inputs.runs-on == 'ubuntu-latest'
+        run: |
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          df -h
+          echo
+          sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
+          sudo apt-get remove --auto-remove android-sdk-platform-tools || true
+          sudo apt-get purge --auto-remove android-sdk-platform-tools || true
+          sudo rm -rf /usr/local/lib/android
+          sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
+          sudo rm -rf /usr/share/dotnet
+          sudo apt-get remove -y '^mono-.*' || true
+          sudo apt-get remove -y '^ghc-.*' || true
+          sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
+          sudo apt-get remove -y 'php.*' || true
+          sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
+          sudo apt-get remove -y '^google-.*' || true
+          sudo apt-get remove -y azure-cli || true
+          sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
+          sudo apt-get remove -y '^gfortran-.*' || true
+          sudo apt-get remove -y microsoft-edge-stable || true
+          sudo apt-get remove -y firefox || true
+          sudo apt-get remove -y powershell || true
+          sudo apt-get remove -y r-base-core || true
+          sudo apt-get autoremove -y
+          sudo apt-get clean
+          echo
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          sudo rm -rfv build || true
+          sudo rm -rf /usr/share/dotnet || true
+          sudo rm -rf /opt/ghc || true
+          sudo rm -rf "/usr/local/share/boost" || true
+          sudo rm -rf "$AGENT_TOOLSDIRECTORY" || true
+          df -h
+
+      - name: Docker meta
+        id: meta
+        if: github.event_name != 'pull_request'
+        uses: docker/metadata-action@v5
+        with:
+          images: |
+            quay.io/go-skynet/local-ai
+            localai/localai
+          tags: |
+            type=ref,event=branch
+            type=semver,pattern={{raw}}
+            type=sha
+          flavor: |
+            latest=${{ inputs.tag-latest }}
+            suffix=${{ inputs.tag-suffix }}
+      - name: Docker meta for PR
+        id: meta_pull_request
+        if: github.event_name == 'pull_request'
+        uses: docker/metadata-action@v5
+        with:
+          images: |
+            ttl.sh/localai-ci-pr-${{ github.event.number }}
+          tags: |
+            type=ref,event=branch
+            type=semver,pattern={{raw}}
+            type=sha
+          flavor: |
+            latest=${{ inputs.tag-latest }}
+            suffix=${{ inputs.tag-suffix }}
+      - name: Docker meta AIO (quay.io)
+        if: inputs.aio != ''
+        id: meta_aio
+        uses: docker/metadata-action@v5
+        with:
+          images: |
+            quay.io/go-skynet/local-ai
+          tags: |
+            type=ref,event=branch
+            type=semver,pattern={{raw}}
+          flavor: |
+            latest=${{ inputs.tag-latest }}
+            suffix=${{ inputs.aio }}
+
+      - name: Docker meta AIO (dockerhub)
+        if: inputs.aio != ''
+        id: meta_aio_dockerhub
+        uses: docker/metadata-action@v5
+        with:
+          images: |
+            localai/localai
+          tags: |
+            type=ref,event=branch
+            type=semver,pattern={{raw}}
+          flavor: |
+            suffix=${{ inputs.aio }}
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@master
+        with:
+          platforms: all
+
+      - name: Set up Docker Buildx
+        id: buildx
+        uses: docker/setup-buildx-action@master
+
+      - name: Login to DockerHub
+        if: github.event_name != 'pull_request'
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.dockerUsername }}
+          password: ${{ secrets.dockerPassword }}
+
+      - name: Login to DockerHub
+        if: github.event_name != 'pull_request'
+        uses: docker/login-action@v3
+        with:
+          registry: quay.io
+          username: ${{ secrets.quayUsername }}
+          password: ${{ secrets.quayPassword }}
+
+      - name: Build and push
+        uses: docker/build-push-action@v6
+        if: github.event_name != 'pull_request'
+        with:
+          builder: ${{ steps.buildx.outputs.name }}
+          # The build-args MUST be an EXACT match between the image cache and other workflow steps that want to use that cache.
+          # This means that even the MAKEFLAGS have to be an EXACT match.
+          # If the build-args are not an EXACT match, it will result in a cache miss, which will require GRPC to be built from scratch.
+          # This is why some build args like GRPC_VERSION and MAKEFLAGS are hardcoded
+          build-args: |
+            BUILD_TYPE=${{ inputs.build-type }}
+            CUDA_MAJOR_VERSION=${{ inputs.cuda-major-version }}
+            CUDA_MINOR_VERSION=${{ inputs.cuda-minor-version }}
+            FFMPEG=${{ inputs.ffmpeg }}
+            IMAGE_TYPE=${{ inputs.image-type }}
+            BASE_IMAGE=${{ inputs.base-image }}
+            GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
+            GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
+            GRPC_VERSION=v1.65.0
+            MAKEFLAGS=${{ inputs.makeflags }}
+            SKIP_DRIVERS=${{ inputs.skip-drivers }}
+          context: .
+          file: ./Dockerfile
+          cache-from: type=gha
+          platforms: ${{ inputs.platforms }}
+          push: ${{ github.event_name != 'pull_request' }}
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+### Start testing image
+      - name: Build and push
+        uses: docker/build-push-action@v6
+        if: github.event_name == 'pull_request'
+        with:
+          builder: ${{ steps.buildx.outputs.name }}
+          # The build-args MUST be an EXACT match between the image cache and other workflow steps that want to use that cache.
+          # This means that even the MAKEFLAGS have to be an EXACT match.
+          # If the build-args are not an EXACT match, it will result in a cache miss, which will require GRPC to be built from scratch.
+          # This is why some build args like GRPC_VERSION and MAKEFLAGS are hardcoded
+          build-args: |
+            BUILD_TYPE=${{ inputs.build-type }}
+            CUDA_MAJOR_VERSION=${{ inputs.cuda-major-version }}
+            CUDA_MINOR_VERSION=${{ inputs.cuda-minor-version }}
+            FFMPEG=${{ inputs.ffmpeg }}
+            IMAGE_TYPE=${{ inputs.image-type }}
+            BASE_IMAGE=${{ inputs.base-image }}
+            GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
+            GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
+            GRPC_VERSION=v1.65.0
+            MAKEFLAGS=${{ inputs.makeflags }}
+            SKIP_DRIVERS=${{ inputs.skip-drivers }}
+          context: .
+          file: ./Dockerfile
+          cache-from: type=gha
+          platforms: ${{ inputs.platforms }}
+          push: true
+          tags: ${{ steps.meta_pull_request.outputs.tags }}
+          labels: ${{ steps.meta_pull_request.outputs.labels }}
+      - name: Testing image
+        if: github.event_name == 'pull_request'
+        run: |
+          echo "Image is available at ttl.sh/localai-ci-pr-${{ github.event.number }}:${{ steps.meta_pull_request.outputs.version }}" >> $GITHUB_STEP_SUMMARY
+## End testing image
+      - name: Build and push AIO image
+        if: inputs.aio != ''
+        uses: docker/build-push-action@v6
+        with:
+          builder: ${{ steps.buildx.outputs.name }}
+          build-args: |
+            BASE_IMAGE=quay.io/go-skynet/local-ai:${{ steps.meta.outputs.version }}
+            MAKEFLAGS=${{ inputs.makeflags }}
+          context: .
+          file: ./Dockerfile.aio
+          platforms: ${{ inputs.platforms }}
+          push: ${{ github.event_name != 'pull_request' }}
+          tags: ${{ steps.meta_aio.outputs.tags }}
+          labels: ${{ steps.meta_aio.outputs.labels }}
+
+      - name: Build and push AIO image (dockerhub)
+        if: inputs.aio != ''
+        uses: docker/build-push-action@v6
+        with:
+          builder: ${{ steps.buildx.outputs.name }}
+          build-args: |
+            BASE_IMAGE=localai/localai:${{ steps.meta.outputs.version }}
+            MAKEFLAGS=${{ inputs.makeflags }}
+          context: .
+          file: ./Dockerfile.aio
+          platforms: ${{ inputs.platforms }}
+          push: ${{ github.event_name != 'pull_request' }}
+          tags: ${{ steps.meta_aio_dockerhub.outputs.tags }}
+          labels: ${{ steps.meta_aio_dockerhub.outputs.labels }}
+
+      - name: Cleanup
+        run: |
+          docker builder prune -f
+          docker system prune --force --volumes --all
+
+      - name: Latest tag
+        # run this on branches, when it is a tag and there is a latest-image defined
+        if: github.event_name != 'pull_request' && inputs.latest-image != ''  && github.ref_type == 'tag'
+        run: |
+          docker pull localai/localai:${{ steps.meta.outputs.version }}
+          docker tag localai/localai:${{ steps.meta.outputs.version }} localai/localai:${{ inputs.latest-image }}
+          docker push localai/localai:${{ inputs.latest-image }}
+          docker pull quay.io/go-skynet/local-ai:${{ steps.meta.outputs.version }}
+          docker tag quay.io/go-skynet/local-ai:${{ steps.meta.outputs.version }} quay.io/go-skynet/local-ai:${{ inputs.latest-image }}
+          docker push quay.io/go-skynet/local-ai:${{ inputs.latest-image }}
+      - name: Latest AIO tag
+        # run this on branches, when it is a tag and there is a latest-image defined
+        if: github.event_name != 'pull_request' && inputs.latest-image-aio != ''  && github.ref_type == 'tag'
+        run: |
+          docker pull localai/localai:${{ steps.meta_aio_dockerhub.outputs.version }}
+          docker tag localai/localai:${{ steps.meta_aio_dockerhub.outputs.version }} localai/localai:${{ inputs.latest-image-aio }}
+          docker push localai/localai:${{ inputs.latest-image-aio }}
+          docker pull quay.io/go-skynet/local-ai:${{ steps.meta_aio.outputs.version }}
+          docker tag quay.io/go-skynet/local-ai:${{ steps.meta_aio.outputs.version }} quay.io/go-skynet/local-ai:${{ inputs.latest-image-aio }}
+          docker push quay.io/go-skynet/local-ai:${{ inputs.latest-image-aio }}
+
+      - name: job summary
+        run: |
+          echo "Built image: ${{ steps.meta.outputs.labels }}" >> $GITHUB_STEP_SUMMARY
+
+      - name: job summary(AIO)
+        if: inputs.aio != ''
+        run: |
+          echo "Built image: ${{ steps.meta_aio.outputs.labels }}" >> $GITHUB_STEP_SUMMARY

.github/workflows/labeler.yml

@@ -0,0 +1,12 @@
+name: "Pull Request Labeler"
+on:
+- pull_request_target
+
+jobs:
+  labeler:
+    permissions:
+      contents: read
+      pull-requests: write
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/labeler@v5

.github/workflows/localaibot_automerge.yml

@@ -0,0 +1,35 @@
+name: LocalAI-bot auto-merge
+on:
+- pull_request_target
+
+permissions:
+  contents: write
+  pull-requests: write
+  packages: read
+
+jobs:
+  dependabot:
+    runs-on: ubuntu-latest
+    if: ${{ github.actor == 'localai-bot' }}
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Approve a PR if not already approved
+        run: |
+          gh pr checkout "$PR_URL"
+            if [ "$(gh pr status --json reviewDecision -q .currentBranch.reviewDecision)" != "APPROVED" ];
+          then
+            gh pr review --approve "$PR_URL"
+          else
+            echo "PR already approved.";
+          fi
+        env:
+          PR_URL: ${{github.event.pull_request.html_url}}
+          GITHUB_TOKEN: ${{secrets.GITHUB_TOKEN}}
+
+      - name: Enable auto-merge for LocalAIBot PRs
+        run: gh pr merge --auto --squash "$PR_URL"
+        env:
+          PR_URL: ${{github.event.pull_request.html_url}}
+          GITHUB_TOKEN: ${{secrets.GITHUB_TOKEN}}

.github/workflows/notify-models.yaml

@@ -0,0 +1,168 @@
+name: Notifications for new models
+on:
+  pull_request:
+     types:
+       - closed
+
+jobs:
+  notify-discord:
+    if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
+    env:
+        MODEL_NAME: gemma-3-12b-it
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        fetch-depth: 0 # needed to checkout all branches for this Action to work
+    - uses: mudler/localai-github-action@v1
+      with:
+        model: 'gemma-3-12b-it' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
+        # Check the PR diff using the current branch and the base branch of the PR
+    - uses: GrantBirki/git-diff-action@v2.8.0
+      id: git-diff-action
+      with:
+            json_diff_file_output: diff.json
+            raw_diff_file_output: diff.txt
+            file_output_only: "true"
+    - name: Summarize
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      id: summarize
+      run: |
+            input="$(cat $DIFF)"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "You are LocalAI-bot. Write a discord message to notify everyone about the new model from the git diff. Make it informal. An example can include: the URL of the model, the name, and a brief description of the model if exists. Also add an hint on how to install it in LocalAI and that can be browsed over https://models.localai.io. For example: local-ai run model_name_here"
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary="$(echo $response | jq -r '.choices[0].message.content')"
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            echo "Summary:"
+            echo "$summary"
+            echo "payload sent"
+            echo "$json_payload"
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+            docker logs --tail 10 local-ai
+    - name: Discord notification
+      env:
+        DISCORD_WEBHOOK: ${{ secrets.DISCORD_WEBHOOK_URL }}
+        DISCORD_USERNAME: "LocalAI-Bot"
+        DISCORD_AVATAR: "https://avatars.githubusercontent.com/u/139863280?v=4"
+      uses: Ilshidur/action-discord@master
+      with:
+        args: ${{ steps.summarize.outputs.message }}
+    - name: Setup tmate session if fails
+      if: ${{ failure() }}
+      uses: mxschmitt/action-tmate@v3.19
+      with:
+        detached: true
+        connect-timeout-seconds: 180
+        limit-access-to-actor: true
+  notify-twitter:
+    if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
+    env:
+        MODEL_NAME: gemma-3-12b-it
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        fetch-depth: 0 # needed to checkout all branches for this Action to work
+    - name: Start LocalAI
+      run: |
+        echo "Starting LocalAI..."
+        docker run -e -ti -d --name local-ai -p 8080:8080 localai/localai:master-ffmpeg-core run --debug $MODEL_NAME
+        until [ "`docker inspect -f {{.State.Health.Status}} local-ai`" == "healthy" ]; do echo "Waiting for container to be ready";  docker logs --tail 10 local-ai; sleep 2; done
+      # Check the PR diff using the current branch and the base branch of the PR
+    - uses: GrantBirki/git-diff-action@v2.8.0
+      id: git-diff-action
+      with:
+            json_diff_file_output: diff.json
+            raw_diff_file_output: diff.txt
+            file_output_only: "true"
+    - name: Summarize
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      id: summarize
+      run: |
+            input="$(cat $DIFF)"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "You are LocalAI-bot. Write a twitter message to notify everyone about the new model from the git diff. Make it informal and really short. An example can include: the name, and a brief description of the model if exists. Also add an hint on how to install it in LocalAI. For example: local-ai run model_name_here"
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary="$(echo $response | jq -r '.choices[0].message.content')"
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            echo "Summary:"
+            echo "$summary"
+            echo "payload sent"
+            echo "$json_payload"
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+            docker logs --tail 10 local-ai
+    - uses: Eomm/why-don-t-you-tweet@v2
+      with:
+        tweet-message: ${{ steps.summarize.outputs.message }}
+      env:
+        # Get your tokens from https://developer.twitter.com/apps
+        TWITTER_CONSUMER_API_KEY: ${{ secrets.TWITTER_APP_KEY }}
+        TWITTER_CONSUMER_API_SECRET: ${{ secrets.TWITTER_APP_SECRET }}
+        TWITTER_ACCESS_TOKEN: ${{ secrets.TWITTER_ACCESS_TOKEN }}
+        TWITTER_ACCESS_TOKEN_SECRET: ${{ secrets.TWITTER_ACCESS_TOKEN_SECRET }}
+    - name: Setup tmate session if fails
+      if: ${{ failure() }}
+      uses: mxschmitt/action-tmate@v3.19
+      with:
+        detached: true
+        connect-timeout-seconds: 180
+        limit-access-to-actor: true

.github/workflows/notify-releases.yaml

@@ -0,0 +1,63 @@
+name: Release notifications
+on:
+  release:
+    types:
+      - published
+
+jobs:
+  notify-discord:
+    runs-on: ubuntu-latest
+    env:
+        RELEASE_BODY: ${{ github.event.release.body }}
+        RELEASE_TITLE: ${{ github.event.release.name }}
+        RELEASE_TAG_NAME: ${{ github.event.release.tag_name }}
+    steps:
+    - uses: mudler/localai-github-action@v1
+      with:
+        model: 'gemma-3-12b-it' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
+    - name: Summarize
+      id: summarize
+      run: |
+            input="$RELEASE_TITLE\b$RELEASE_BODY"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "Write a discord message with a bullet point summary of the release notes."
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI API
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary=$(echo $response | jq -r '.choices[0].message.content')
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+    - name: Discord notification
+      env:
+        DISCORD_WEBHOOK: ${{ secrets.DISCORD_WEBHOOK_URL_RELEASE }}
+        DISCORD_USERNAME: "LocalAI-Bot"
+        DISCORD_AVATAR: "https://avatars.githubusercontent.com/u/139863280?v=4"
+      uses: Ilshidur/action-discord@master
+      with:
+        args: ${{ steps.summarize.outputs.message }}

.github/workflows/prlint.yaml

@@ -0,0 +1,28 @@
+name: Check PR style
+
+on:
+  pull_request_target:
+    types:
+      - opened
+      - reopened
+      - edited
+      - synchronize
+
+jobs:
+  title-lint:
+    runs-on: ubuntu-latest
+    permissions:
+      statuses: write
+    steps:
+      - uses: aslafy-z/conventional-pr-title-action@v3
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+#  check-pr-description:
+#    runs-on: ubuntu-latest
+#    steps:
+#      - uses: actions/checkout@v2
+#      - uses: jadrol/pr-description-checker-action@v1.0.0
+#        id: description-checker
+#        with:
+#          repo-token: ${{ secrets.GITHUB_TOKEN }}
+#          exempt-labels: no qa

.github/workflows/release.yaml

@@ -0,0 +1,324 @@
+name: Build and Release
+
+on:
+  push:
+    branches:
+      - master
+    tags:
+      - 'v*'
+  pull_request:
+
+env:
+  GRPC_VERSION: v1.65.0
+
+permissions:
+  contents: write
+
+concurrency:
+  group: ci-releases-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
+jobs:
+
+  build-linux-arm:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - uses: actions/setup-go@v5
+        with:
+          go-version: '1.21.x'
+          cache: false
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install build-essential ffmpeg protobuf-compiler ccache upx-ucl gawk
+          sudo apt-get install -qy binutils-aarch64-linux-gnu gcc-aarch64-linux-gnu g++-aarch64-linux-gnu libgmock-dev
+      - name: Install CUDA Dependencies
+        run: |
+          curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/cross-linux-aarch64/cuda-keyring_1.1-1_all.deb
+          sudo dpkg -i cuda-keyring_1.1-1_all.deb
+          sudo apt-get update
+          sudo apt-get install -y cuda-cross-aarch64 cuda-nvcc-cross-aarch64-${CUDA_VERSION} libcublas-cross-aarch64-${CUDA_VERSION}
+        env:
+          CUDA_VERSION: 12-4
+      - name: Cache grpc
+        id: cache-grpc
+        uses: actions/cache@v4
+        with:
+          path: grpc
+          key: ${{ runner.os }}-arm-grpc-${{ env.GRPC_VERSION }}
+      - name: Build grpc
+        if: steps.cache-grpc.outputs.cache-hit != 'true'
+        run: |
+
+          git clone --recurse-submodules -b ${{ env.GRPC_VERSION }} --depth 1 --shallow-submodules https://github.com/grpc/grpc && \
+          cd grpc && sed -i "216i\  TESTONLY" "third_party/abseil-cpp/absl/container/CMakeLists.txt" && mkdir -p cmake/build && \
+          cd cmake/build && cmake -DgRPC_INSTALL=ON \
+            -DgRPC_BUILD_TESTS=OFF \
+            ../.. && sudo make --jobs 5 --output-sync=target
+      - name: Install gRPC
+        run: |
+          GNU_HOST=aarch64-linux-gnu
+          C_COMPILER_ARM_LINUX=$GNU_HOST-gcc
+          CXX_COMPILER_ARM_LINUX=$GNU_HOST-g++
+
+          CROSS_TOOLCHAIN=/usr/$GNU_HOST
+          CROSS_STAGING_PREFIX=$CROSS_TOOLCHAIN/stage
+          CMAKE_CROSS_TOOLCHAIN=/tmp/arm.toolchain.cmake
+
+          # https://cmake.org/cmake/help/v3.13/manual/cmake-toolchains.7.html#cross-compiling-for-linux
+          echo "set(CMAKE_SYSTEM_NAME Linux)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_SYSTEM_PROCESSOR arm)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_STAGING_PREFIX $CROSS_STAGING_PREFIX)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_SYSROOT ${CROSS_TOOLCHAIN}/sysroot)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_C_COMPILER /usr/bin/$C_COMPILER_ARM_LINUX)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_CXX_COMPILER /usr/bin/$CXX_COMPILER_ARM_LINUX)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_FIND_ROOT_PATH_MODE_PROGRAM NEVER)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_FIND_ROOT_PATH_MODE_LIBRARY ONLY)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_FIND_ROOT_PATH_MODE_INCLUDE ONLY)" >> $CMAKE_CROSS_TOOLCHAIN && \
+            echo "set(CMAKE_FIND_ROOT_PATH_MODE_PACKAGE ONLY)" >> $CMAKE_CROSS_TOOLCHAIN
+          GRPC_DIR=$PWD/grpc
+          cd grpc && cd cmake/build && sudo make --jobs 5 --output-sync=target install && \
+          GRPC_CROSS_BUILD_DIR=$GRPC_DIR/cmake/cross_build && \
+          mkdir -p $GRPC_CROSS_BUILD_DIR && \
+          cd $GRPC_CROSS_BUILD_DIR && \
+          cmake -DCMAKE_TOOLCHAIN_FILE=$CMAKE_CROSS_TOOLCHAIN \
+            -DCMAKE_BUILD_TYPE=Release \
+            -DCMAKE_INSTALL_PREFIX=$CROSS_TOOLCHAIN/grpc_install \
+            ../.. && \
+          sudo make -j`nproc` install
+      - name: Build
+        id: build
+        run: |
+          GNU_HOST=aarch64-linux-gnu
+          C_COMPILER_ARM_LINUX=$GNU_HOST-gcc
+          CXX_COMPILER_ARM_LINUX=$GNU_HOST-g++
+
+          CROSS_TOOLCHAIN=/usr/$GNU_HOST
+          CROSS_STAGING_PREFIX=$CROSS_TOOLCHAIN/stage
+          CMAKE_CROSS_TOOLCHAIN=/tmp/arm.toolchain.cmake
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          export PATH=$PATH:$GOPATH/bin
+          export PATH=/usr/local/cuda/bin:$PATH
+          sudo rm -rf /usr/aarch64-linux-gnu/lib/libstdc++.so.6
+          sudo cp -rf /usr/aarch64-linux-gnu/lib/libstdc++.so* /usr/aarch64-linux-gnu/lib/libstdc++.so.6
+          sudo cp /usr/aarch64-linux-gnu/lib/ld-linux-aarch64.so.1 ld.so
+          BACKEND_LIBS="./grpc/cmake/cross_build/third_party/re2/libre2.a ./grpc/cmake/cross_build/libgrpc.a ./grpc/cmake/cross_build/libgrpc++.a ./grpc/cmake/cross_build/third_party/protobuf/libprotobuf.a /usr/aarch64-linux-gnu/lib/libc.so.6 /usr/aarch64-linux-gnu/lib/libstdc++.so.6 /usr/aarch64-linux-gnu/lib/libgomp.so.1 /usr/aarch64-linux-gnu/lib/libm.so.6 /usr/aarch64-linux-gnu/lib/libgcc_s.so.1 /usr/aarch64-linux-gnu/lib/libdl.so.2 /usr/aarch64-linux-gnu/lib/libpthread.so.0 ./ld.so" \
+          GOOS=linux \
+          GOARCH=arm64 \
+          CMAKE_ARGS="-DProtobuf_INCLUDE_DIRS=$CROSS_STAGING_PREFIX/include -DProtobuf_DIR=$CROSS_STAGING_PREFIX/lib/cmake/protobuf -DgRPC_DIR=$CROSS_STAGING_PREFIX/lib/cmake/grpc -DCMAKE_TOOLCHAIN_FILE=$CMAKE_CROSS_TOOLCHAIN -DCMAKE_C_COMPILER=aarch64-linux-gnu-gcc -DCMAKE_CXX_COMPILER=aarch64-linux-gnu-g++" make dist-cross-linux-arm64
+      - uses: actions/upload-artifact@v4
+        with:
+          name: LocalAI-linux-arm64
+          path: release/
+      - name: Release
+        uses: softprops/action-gh-release@v2
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: |
+            release/*
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.19
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
+  build-linux:
+    runs-on: arc-runner-set
+    steps:
+      - name: Force Install GIT latest
+        run: |
+          sudo apt-get update \
+          && sudo apt-get install -y software-properties-common \
+          && sudo apt-get update \
+          && sudo add-apt-repository -y ppa:git-core/ppa \
+          && sudo apt-get update \
+          && sudo apt-get install -y git
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - uses: actions/setup-go@v5
+        with:
+          go-version: '1.21.x'
+          cache: false
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y wget curl build-essential ffmpeg protobuf-compiler ccache upx-ucl gawk cmake libgmock-dev
+      - name: Intel Dependencies
+        run: |
+          wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | sudo tee /usr/share/keyrings/oneapi-archive-keyring.gpg > /dev/null
+          echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt.repos.intel.com/oneapi all main" | sudo tee /etc/apt/sources.list.d/oneAPI.list
+          sudo apt update
+          sudo apt install -y intel-basekit
+      - name: Install CUDA Dependencies
+        run: |
+          curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
+          sudo dpkg -i cuda-keyring_1.1-1_all.deb
+          sudo apt-get update
+          sudo apt-get install -y cuda-nvcc-${CUDA_VERSION} libcublas-dev-${CUDA_VERSION}
+        env:
+          CUDA_VERSION: 12-5
+      - name: "Install Hipblas"
+        env:
+          ROCM_VERSION: "6.1"
+          AMDGPU_VERSION: "6.1"
+        run: |
+            set -ex
+
+            sudo apt-get update
+            sudo DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends ca-certificates curl libnuma-dev gnupg
+
+            curl -sL https://repo.radeon.com/rocm/rocm.gpg.key | sudo apt-key add -
+
+            printf "deb [arch=amd64] https://repo.radeon.com/rocm/apt/$ROCM_VERSION/ jammy main" | sudo tee /etc/apt/sources.list.d/rocm.list
+
+            printf "deb [arch=amd64] https://repo.radeon.com/amdgpu/$AMDGPU_VERSION/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+            printf 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
+            sudo apt-get update
+
+            sudo DEBIAN_FRONTEND=noninteractive apt-get install -y \
+                hipblas-dev rocm-dev \
+                rocblas-dev
+
+            sudo apt-get clean
+            sudo rm -rf /var/lib/apt/lists/*
+            sudo ldconfig
+      - name: Cache grpc
+        id: cache-grpc
+        uses: actions/cache@v4
+        with:
+          path: grpc
+          key: ${{ runner.os }}-grpc-${{ env.GRPC_VERSION }}
+      - name: Build grpc
+        if: steps.cache-grpc.outputs.cache-hit != 'true'
+        run: |
+          git clone --recurse-submodules -b ${{ env.GRPC_VERSION }} --depth 1 --shallow-submodules https://github.com/grpc/grpc && \
+          cd grpc && sed -i "216i\  TESTONLY" "third_party/abseil-cpp/absl/container/CMakeLists.txt" && mkdir -p cmake/build && \
+          cd cmake/build && cmake -DgRPC_INSTALL=ON \
+            -DgRPC_BUILD_TESTS=OFF \
+            ../.. && sudo make --jobs 5 --output-sync=target
+      - name: Install gRPC
+        run: |
+          cd grpc && cd cmake/build && sudo make --jobs 5 --output-sync=target install
+      # BACKEND_LIBS needed for gpu-workload: /opt/intel/oneapi/*/lib/libiomp5.so /opt/intel/oneapi/*/lib/libmkl_core.so /opt/intel/oneapi/*/lib/libmkl_core.so.2 /opt/intel/oneapi/*/lib/libmkl_intel_ilp64.so /opt/intel/oneapi/*/lib/libmkl_intel_ilp64.so.2 /opt/intel/oneapi/*/lib/libmkl_sycl_blas.so /opt/intel/oneapi/*/lib/libmkl_sycl_blas.so.4 /opt/intel/oneapi/*/lib/libmkl_tbb_thread.so /opt/intel/oneapi/*/lib/libmkl_tbb_thread.so.2 /opt/intel/oneapi/*/lib/libsycl.so /opt/intel/oneapi/*/lib/libsycl.so.7 /opt/intel/oneapi/*/lib/libsycl.so.7.1.0 /opt/rocm-*/lib/libamdhip64.so /opt/rocm-*/lib/libamdhip64.so.5 /opt/rocm-*/lib/libamdhip64.so.6 /opt/rocm-*/lib/libamdhip64.so.6.1.60100 /opt/rocm-*/lib/libhipblas.so /opt/rocm-*/lib/libhipblas.so.2 /opt/rocm-*/lib/libhipblas.so.2.1.60100 /opt/rocm-*/lib/librocblas.so /opt/rocm-*/lib/librocblas.so.4 /opt/rocm-*/lib/librocblas.so.4.1.60100 /usr/lib/x86_64-linux-gnu/libstdc++.so.6 /usr/lib/x86_64-linux-gnu/libOpenCL.so.1 /usr/lib/x86_64-linux-gnu/libOpenCL.so.1.0.0 /usr/lib/x86_64-linux-gnu/libm.so.6 /usr/lib/x86_64-linux-gnu/libgcc_s.so.1 /usr/lib/x86_64-linux-gnu/libc.so.6 /usr/lib/x86_64-linux-gnu/librt.so.1 /usr/local/cuda-*/targets/x86_64-linux/lib/libcublas.so /usr/local/cuda-*/targets/x86_64-linux/lib/libcublasLt.so /usr/local/cuda-*/targets/x86_64-linux/lib/libcudart.so /usr/local/cuda-*/targets/x86_64-linux/lib/stubs/libcuda.so
+      - name: Build
+        id: build
+        run: |
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          export PATH=$PATH:$GOPATH/bin
+          export PATH=/usr/local/cuda/bin:$PATH
+          export PATH=/opt/rocm/bin:$PATH
+          source /opt/intel/oneapi/setvars.sh
+          sudo cp /lib64/ld-linux-x86-64.so.2 ld.so
+          BACKEND_LIBS="./ld.so ./sources/go-piper/piper/build/fi/lib/libfmt.a ./sources/go-piper/piper-phonemize/pi/lib/libonnxruntime.so.1.14.1 ./sources/go-piper/piper-phonemize/pi/src/libespeak-ng/libespeak-ng.so /usr/lib/x86_64-linux-gnu/libdl.so.2 /usr/lib/x86_64-linux-gnu/librt.so.1 /usr/lib/x86_64-linux-gnu/libpthread.so.0 ./sources/go-piper/piper-phonemize/pi/lib/libpiper_phonemize.so.1 ./sources/go-piper/piper/build/si/lib/libspdlog.a ./sources/go-piper/espeak/ei/lib/libucd.so" \
+          make -j4 dist
+      - uses: actions/upload-artifact@v4
+        with:
+          name: LocalAI-linux
+          path: release/
+      - name: Release
+        uses: softprops/action-gh-release@v2
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: |
+            release/*
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.19
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
+
+
+  build-macOS-x86_64:
+    runs-on: macos-13
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - uses: actions/setup-go@v5
+        with:
+          go-version: '1.21.x'
+          cache: false
+      - name: Dependencies
+        run: |
+          brew install protobuf grpc
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
+      - name: Build
+        id: build
+        run: |
+          export C_INCLUDE_PATH=/usr/local/include
+          export CPLUS_INCLUDE_PATH=/usr/local/include
+          export PATH=$PATH:$GOPATH/bin
+          export SKIP_GRPC_BACKEND=backend-assets/grpc/whisper
+          make dist
+      - uses: actions/upload-artifact@v4
+        with:
+          name: LocalAI-MacOS-x86_64
+          path: release/
+      - name: Release
+        uses: softprops/action-gh-release@v2
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: |
+            release/*
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.19
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
+
+  build-macOS-arm64:
+    runs-on: macos-14
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - uses: actions/setup-go@v5
+        with:
+          go-version: '1.21.x'
+          cache: false
+      - name: Dependencies
+        run: |
+          brew install protobuf grpc libomp llvm
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+      - name: Build
+        id: build
+        run: |
+          export C_INCLUDE_PATH=/usr/local/include
+          export CPLUS_INCLUDE_PATH=/usr/local/include
+          export PATH=$PATH:$GOPATH/bin
+          export CC=/opt/homebrew/opt/llvm/bin/clang
+          make dist
+      - uses: actions/upload-artifact@v4
+        with:
+          name: LocalAI-MacOS-arm64
+          path: release/
+      - name: Release
+        uses: softprops/action-gh-release@v2
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: |
+            release/*
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.19
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true

.github/workflows/release.yml.disabled

@@ -1,26 +0,0 @@
-name: goreleaser
-
-on:
-  push:
-    tags:
-      - 'v*'
-
-jobs:
-  goreleaser:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          fetch-depth: 0
-      - name: Set up Go
-        uses: actions/setup-go@v3
-        with:
-          go-version: 1.18
-      - name: Run GoReleaser
-        uses: goreleaser/goreleaser-action@v4
-        with:
-          version: latest
-          args: release --clean
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/secscan.yaml

@@ -0,0 +1,30 @@
+name: "Security Scan"
+
+# Run workflow each time code is pushed to your repository and on a schedule.
+# The scheduled workflow runs every at 00:00 on Sunday UTC time.
+on:
+  push:
+  schedule:
+  - cron: '0 0 * * 0'
+
+jobs:
+  tests:
+    runs-on: ubuntu-latest
+    env:
+      GO111MODULE: on
+    steps:
+      - name: Checkout Source
+        uses: actions/checkout@v4
+        if: ${{ github.actor != 'dependabot[bot]' }}
+      - name: Run Gosec Security Scanner
+        if: ${{ github.actor != 'dependabot[bot]' }}
+        uses: securego/gosec@v2.22.3
+        with:
+          # we let the report trigger content trigger a failure using the GitHub Security features.
+          args: '-no-fail -fmt sarif -out results.sarif ./...'
+      - name: Upload SARIF file
+        if: ${{ github.actor != 'dependabot[bot]' }}
+        uses: github/codeql-action/upload-sarif@v3
+        with:
+          # Path to SARIF file relative to the root of the repository
+          sarif_file: results.sarif

.github/workflows/test-extra.yml

@@ -0,0 +1,207 @@
+---
+name: 'Tests extras backends'
+
+on:
+  pull_request:
+  push:
+    branches:
+      - master
+    tags:
+      - '*'
+
+concurrency:
+  group: ci-tests-extra-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
+jobs:
+  tests-transformers:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install build-essential ffmpeg
+          # Install UV
+          curl -LsSf https://astral.sh/uv/install.sh | sh
+          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
+          sudo apt-get install -y libopencv-dev
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
+
+      - name: Test transformers
+        run: |
+           make --jobs=5 --output-sync=target -C backend/python/transformers
+           make --jobs=5 --output-sync=target -C backend/python/transformers test
+  tests-rerankers:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install build-essential ffmpeg
+          # Install UV
+          curl -LsSf https://astral.sh/uv/install.sh | sh
+          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
+          sudo apt-get install -y libopencv-dev
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
+
+      - name: Test rerankers
+        run: |
+           make --jobs=5 --output-sync=target -C backend/python/rerankers
+           make --jobs=5 --output-sync=target -C backend/python/rerankers test
+
+  tests-diffusers:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y build-essential ffmpeg
+          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
+          sudo apt-get install -y libopencv-dev
+          # Install UV
+          curl -LsSf https://astral.sh/uv/install.sh | sh
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
+      - name: Test diffusers
+        run: |
+          make --jobs=5 --output-sync=target -C backend/python/diffusers
+          make --jobs=5 --output-sync=target -C backend/python/diffusers test
+
+  # tests-transformers-musicgen:
+  #   runs-on: ubuntu-latest
+  #   steps:
+  #     - name: Clone
+  #       uses: actions/checkout@v4
+  #       with:
+  #         submodules: true
+  #     - name: Dependencies
+  #       run: |
+  #         sudo apt-get update
+  #         sudo apt-get install build-essential ffmpeg
+  #         # Install UV
+  #         curl -LsSf https://astral.sh/uv/install.sh | sh
+  #         sudo apt-get install -y ca-certificates cmake curl patch python3-pip
+  #         sudo apt-get install -y libopencv-dev
+  #         pip install --user --no-cache-dir grpcio-tools==1.64.1
+
+  #     - name: Test transformers-musicgen
+  #       run: |
+  #          make --jobs=5 --output-sync=target -C backend/python/transformers-musicgen
+  #          make --jobs=5 --output-sync=target -C backend/python/transformers-musicgen test
+
+  # tests-bark:
+  #   runs-on: ubuntu-latest
+  #   steps:
+  #     - name: Release space from worker
+  #       run: |
+  #           echo "Listing top largest packages"
+  #           pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+  #           head -n 30 <<< "${pkgs}"
+  #           echo
+  #           df -h
+  #           echo
+  #           sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
+  #           sudo apt-get remove --auto-remove android-sdk-platform-tools || true
+  #           sudo apt-get purge --auto-remove android-sdk-platform-tools || true
+  #           sudo rm -rf /usr/local/lib/android
+  #           sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
+  #           sudo rm -rf /usr/share/dotnet
+  #           sudo apt-get remove -y '^mono-.*' || true
+  #           sudo apt-get remove -y '^ghc-.*' || true
+  #           sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
+  #           sudo apt-get remove -y 'php.*' || true
+  #           sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
+  #           sudo apt-get remove -y '^google-.*' || true
+  #           sudo apt-get remove -y azure-cli || true
+  #           sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
+  #           sudo apt-get remove -y '^gfortran-.*' || true
+  #           sudo apt-get remove -y microsoft-edge-stable || true
+  #           sudo apt-get remove -y firefox || true
+  #           sudo apt-get remove -y powershell || true
+  #           sudo apt-get remove -y r-base-core || true
+  #           sudo apt-get autoremove -y
+  #           sudo apt-get clean
+  #           echo
+  #           echo "Listing top largest packages"
+  #           pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+  #           head -n 30 <<< "${pkgs}"
+  #           echo
+  #           sudo rm -rfv build || true
+  #           sudo rm -rf /usr/share/dotnet || true
+  #           sudo rm -rf /opt/ghc || true
+  #           sudo rm -rf "/usr/local/share/boost" || true
+  #           sudo rm -rf "$AGENT_TOOLSDIRECTORY" || true
+  #           df -h
+  #     - name: Clone
+  #       uses: actions/checkout@v4
+  #       with:
+  #         submodules: true
+  #     - name: Dependencies
+  #       run: |
+  #         sudo apt-get update
+  #         sudo apt-get install build-essential ffmpeg
+  #         # Install UV
+  #         curl -LsSf https://astral.sh/uv/install.sh | sh
+  #         sudo apt-get install -y ca-certificates cmake curl patch python3-pip
+  #         sudo apt-get install -y libopencv-dev
+  #         pip install --user --no-cache-dir grpcio-tools==1.64.1
+
+  #     - name: Test bark
+  #       run: |
+  #          make --jobs=5 --output-sync=target -C backend/python/bark
+  #          make --jobs=5 --output-sync=target -C backend/python/bark test
+
+
+  # Below tests needs GPU. Commented out for now
+  # TODO: Re-enable as soon as we have GPU nodes
+  # tests-vllm:
+  #   runs-on: ubuntu-latest
+  #   steps:
+  #     - name: Clone
+  #       uses: actions/checkout@v4
+  #       with:
+  #         submodules: true
+  #     - name: Dependencies
+  #       run: |
+  #         sudo apt-get update
+  #         sudo apt-get install build-essential ffmpeg
+  #         # Install UV
+  #         curl -LsSf https://astral.sh/uv/install.sh | sh
+  #         sudo apt-get install -y ca-certificates cmake curl patch python3-pip
+  #         sudo apt-get install -y libopencv-dev
+  #         pip install --user --no-cache-dir grpcio-tools==1.64.1
+  #     - name: Test vllm
+  #       run: |
+  #          make --jobs=5 --output-sync=target -C backend/python/vllm
+  #          make --jobs=5 --output-sync=target -C backend/python/vllm test
+
+  tests-coqui:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install build-essential ffmpeg
+          sudo apt-get install -y ca-certificates cmake curl patch espeak espeak-ng python3-pip
+          # Install UV
+          curl -LsSf https://astral.sh/uv/install.sh | sh
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
+      - name: Test coqui
+        run: |
+          make --jobs=5 --output-sync=target -C backend/python/coqui
+          make --jobs=5 --output-sync=target -C backend/python/coqui test

.github/workflows/test.yml

@@ -9,36 +9,231 @@ on:
     tags:
       - '*'
 
-jobs:
-  ubuntu-latest:
-    runs-on: ubuntu-latest
+env:
+  GRPC_VERSION: v1.65.0
 
+concurrency:
+  group: ci-tests-${{ github.head_ref || github.ref }}-${{ github.repository }}
+  cancel-in-progress: true
+
+jobs:
+  tests-linux:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        go-version: ['1.21.x']
     steps:
+      - name: Release space from worker
+        run: |
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          df -h
+          echo
+          sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
+          sudo apt-get remove --auto-remove android-sdk-platform-tools || true
+          sudo apt-get purge --auto-remove android-sdk-platform-tools || true
+          sudo rm -rf /usr/local/lib/android
+          sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
+          sudo rm -rf /usr/share/dotnet
+          sudo apt-get remove -y '^mono-.*' || true
+          sudo apt-get remove -y '^ghc-.*' || true
+          sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
+          sudo apt-get remove -y 'php.*' || true
+          sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
+          sudo apt-get remove -y '^google-.*' || true
+          sudo apt-get remove -y azure-cli || true
+          sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
+          sudo apt-get remove -y '^gfortran-.*' || true
+          sudo apt-get autoremove -y
+          sudo apt-get clean
+          echo
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          sudo rm -rfv build || true
+          df -h
       - name: Clone
-        uses: actions/checkout@v3
-        with: 
+        uses: actions/checkout@v4
+        with:
           submodules: true
+      - name: Setup Go ${{ matrix.go-version }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ matrix.go-version }}
+          cache: false
+      # You can test your matrix by printing the current Go version
+      - name: Display Go version
+        run: go version
       - name: Dependencies
         run: |
           sudo apt-get update
-          sudo apt-get install build-essential ffmpeg
+          sudo apt-get install build-essential ccache upx-ucl curl ffmpeg
+          sudo apt-get install -y libgmock-dev
+          curl https://repo.anaconda.com/pkgs/misc/gpgkeys/anaconda.asc | gpg --dearmor > conda.gpg && \
+             sudo install -o root -g root -m 644 conda.gpg /usr/share/keyrings/conda-archive-keyring.gpg && \
+             gpg --keyring /usr/share/keyrings/conda-archive-keyring.gpg --no-default-keyring --fingerprint 34161F5BF5EB1D4BFBBB8F0A8AEB4F8B29D82806 && \
+             sudo /bin/bash -c 'echo "deb [arch=amd64 signed-by=/usr/share/keyrings/conda-archive-keyring.gpg] https://repo.anaconda.com/pkgs/misc/debrepo/conda stable main" > /etc/apt/sources.list.d/conda.list' && \
+             sudo /bin/bash -c 'echo "deb [arch=amd64 signed-by=/usr/share/keyrings/conda-archive-keyring.gpg] https://repo.anaconda.com/pkgs/misc/debrepo/conda stable main" | tee -a /etc/apt/sources.list.d/conda.list' && \
+             sudo apt-get update && \
+             sudo apt-get install -y conda
+          # Install UV
+          curl -LsSf https://astral.sh/uv/install.sh | sh
+          sudo apt-get install -y ca-certificates cmake patch python3-pip unzip
+          sudo apt-get install -y libopencv-dev
+
+          curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v26.1/protoc-26.1-linux-x86_64.zip -o protoc.zip && \
+          unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
+          rm protoc.zip
+
+          curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
+          sudo dpkg -i cuda-keyring_1.1-1_all.deb
+          sudo apt-get update
+          sudo apt-get install -y cuda-nvcc-${CUDA_VERSION} libcublas-dev-${CUDA_VERSION}
+          export CUDACXX=/usr/local/cuda/bin/nvcc
+
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+
+          # The python3-grpc-tools package in 22.04 is too old
+          pip install --user grpcio-tools
+
+          make -C backend/python/transformers
+
+          # Pre-build piper before we start tests in order to have shared libraries in place
+          make sources/go-piper && \
+          GO_TAGS="tts" make -C sources/go-piper piper.o && \
+          sudo cp -rfv sources/go-piper/piper-phonemize/pi/lib/. /usr/lib/
+        env:
+          CUDA_VERSION: 12-4
+      - name: Cache grpc
+        id: cache-grpc
+        uses: actions/cache@v4
+        with:
+          path: grpc
+          key: ${{ runner.os }}-grpc-${{ env.GRPC_VERSION }}
+      - name: Build grpc
+        if: steps.cache-grpc.outputs.cache-hit != 'true'
+        run: |
+          git clone --recurse-submodules -b ${{ env.GRPC_VERSION }} --depth 1 --jobs 5 --shallow-submodules https://github.com/grpc/grpc && \
+          cd grpc && sed -i "216i\  TESTONLY" "third_party/abseil-cpp/absl/container/CMakeLists.txt" && mkdir -p cmake/build && cd cmake/build && \
+          cmake -DgRPC_INSTALL=ON \
+            -DgRPC_BUILD_TESTS=OFF \
+            ../.. && sudo make --jobs 5
+      - name: Install gRPC
+        run: |
+          cd grpc && cd cmake/build && sudo make --jobs 5 install
       - name: Test
         run: |
-          make test
-
-  macOS-latest:
-    runs-on: macOS-latest
+          PATH="$PATH:/root/go/bin" GO_TAGS="tts" make --jobs 5 --output-sync=target test
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.19
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
 
+  tests-aio-container:
+    runs-on: ubuntu-latest
     steps:
+      - name: Release space from worker
+        run: |
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          df -h
+          echo
+          sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
+          sudo apt-get remove --auto-remove android-sdk-platform-tools || true
+          sudo apt-get purge --auto-remove android-sdk-platform-tools || true
+          sudo rm -rf /usr/local/lib/android
+          sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
+          sudo rm -rf /usr/share/dotnet
+          sudo apt-get remove -y '^mono-.*' || true
+          sudo apt-get remove -y '^ghc-.*' || true
+          sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
+          sudo apt-get remove -y 'php.*' || true
+          sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
+          sudo apt-get remove -y '^google-.*' || true
+          sudo apt-get remove -y azure-cli || true
+          sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
+          sudo apt-get remove -y '^gfortran-.*' || true
+          sudo apt-get autoremove -y
+          sudo apt-get clean
+          echo
+          echo "Listing top largest packages"
+          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+          head -n 30 <<< "${pkgs}"
+          echo
+          sudo rm -rfv build || true
+          df -h
       - name: Clone
-        uses: actions/checkout@v3
-        with: 
+        uses: actions/checkout@v4
+        with:
           submodules: true
-
       - name: Dependencies
         run: |
-          brew update
-          brew install sdl2 ffmpeg
+          # Install protoc
+          curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v26.1/protoc-26.1-linux-x86_64.zip -o protoc.zip && \
+          unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
+          rm protoc.zip
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          PATH="$PATH:$HOME/go/bin" make protogen-go
+      - name: Build images
+        run: |
+          docker build --build-arg FFMPEG=true --build-arg IMAGE_TYPE=extras --build-arg EXTRA_BACKENDS=rerankers --build-arg MAKEFLAGS="--jobs=5 --output-sync=target" -t local-ai:tests -f Dockerfile .
+          BASE_IMAGE=local-ai:tests DOCKER_AIO_IMAGE=local-ai-aio:test make docker-aio
       - name: Test
         run: |
-          make test
+            PATH="$PATH:$HOME/go/bin" LOCALAI_MODELS_DIR=$PWD/models LOCALAI_IMAGE_TAG=test LOCALAI_IMAGE=local-ai-aio \
+            make run-e2e-aio
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.19
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
+
+  tests-apple:
+    runs-on: macOS-14
+    strategy:
+      matrix:
+        go-version: ['1.21.x']
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - name: Setup Go ${{ matrix.go-version }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ matrix.go-version }}
+          cache: false
+      # You can test your matrix by printing the current Go version
+      - name: Display Go version
+        run: go version
+      - name: Dependencies
+        run: |
+          brew install protobuf grpc make protoc-gen-go protoc-gen-go-grpc libomp llvm
+          pip install --user --no-cache-dir grpcio-tools
+      - name: Test
+        run: |
+          export C_INCLUDE_PATH=/usr/local/include
+          export CPLUS_INCLUDE_PATH=/usr/local/include
+          export CC=/opt/homebrew/opt/llvm/bin/clang
+          # Used to run the newer GNUMake version from brew that supports --output-sync
+          export PATH="/opt/homebrew/opt/make/libexec/gnubin:$PATH"
+          BUILD_TYPE="GITHUB_CI_HAS_BROKEN_METAL" CMAKE_ARGS="-DGGML_F16C=OFF -DGGML_AVX512=OFF -DGGML_AVX2=OFF -DGGML_FMA=OFF" make --jobs 4 --output-sync=target test
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.19
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true

.github/workflows/update_swagger.yaml

@@ -0,0 +1,37 @@
+name: Update swagger
+on:
+  schedule:
+    - cron: 0 20 * * *
+  workflow_dispatch:
+jobs:
+  swagger:
+    strategy:
+      fail-fast: false
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-go@v5
+        with:
+          go-version: 'stable'
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install protobuf-compiler
+      - run: |
+          go install github.com/swaggo/swag/cmd/swag@latest
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+      - name: Bump swagger 🔧
+        run: |
+          make protogen-go swagger
+      - name: Create Pull Request
+        uses: peter-evans/create-pull-request@v7
+        with:
+          token: ${{ secrets.UPDATE_BOT_TOKEN }}
+          push-to-fork: ci-forks/LocalAI
+          commit-message: 'feat(swagger): update swagger'
+          title: 'feat(swagger): update swagger'
+          branch: "update/swagger"
+          body:  Update swagger
+          signoff: true
+

.github/workflows/yaml-check.yml

@@ -0,0 +1,18 @@
+name: 'Yamllint GitHub Actions'
+on:
+  - pull_request
+jobs:
+  yamllint:
+    name: 'Yamllint'
+    runs-on: ubuntu-latest
+    steps:
+      - name: 'Checkout'
+        uses: actions/checkout@master
+      - name: 'Yamllint'
+        uses: karancode/yamllint-github-action@master
+        with:
+          yamllint_file_or_dir: 'gallery'
+          yamllint_strict: false
+          yamllint_comment: true
+        env:
+          GITHUB_ACCESS_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.gitignore

@@ -1,20 +1,59 @@
 # go-llama build artifacts
-go-llama
-gpt4all
-go-stable-diffusion
+/sources/
+__pycache__/
+*.a
+*.o
+get-sources
+prepare-sources
+/backend/cpp/llama/grpc-server
+/backend/cpp/llama/llama.cpp
+/backend/cpp/llama-*
+
+*.log
+
+go-ggml-transformers
 go-gpt2
-go-rwkv
 whisper.cpp
+/bloomz
+go-bert
 
 # LocalAI build binary
 LocalAI
 local-ai
 # prevent above rules from omitting the helm chart
 !charts/*
+# prevent above rules from omitting the api/localai folder
+!api/localai
+!core/**/localai
 
 # Ignore models
 models/*
 test-models/
+test-dir/
+
+release/
 
 # just in case
 .DS_Store
+.idea
+
+# Generated during build
+backend-assets/*
+!backend-assets/.keep
+prepare
+/ggml-metal.metal
+docs/static/gallery.html
+
+# Protobuf generated files
+*.pb.go
+*pb2.py
+*pb2_grpc.py
+
+# SonarQube
+.scannerwork
+
+# backend virtual environments
+**/venv
+
+# per-developer customization files for the development container
+.devcontainer/customization/*

.gitmodules

@@ -0,0 +1,6 @@
+[submodule "docs/themes/hugo-theme-relearn"]
+	path = docs/themes/hugo-theme-relearn
+	url = https://github.com/McShelby/hugo-theme-relearn.git
+[submodule "docs/themes/lotusdocs"]
+	path = docs/themes/lotusdocs
+	url = https://github.com/colinwilson/lotusdocs

.goreleaser.yaml

@@ -1,15 +0,0 @@
-# Make sure to check the documentation at http://goreleaser.com
-project_name: local-ai
-builds:
-  - ldflags:
-      - -w -s
-    env:
-      - CGO_ENABLED=0
-    goos:
-      - linux
-      - darwin
-      - windows
-    goarch:
-      - amd64
-      - arm64
-    binary: '{{ .ProjectName }}'

.vscode/extensions.json

@@ -0,0 +1,5 @@
+{
+    "recommendations": [
+        "golang.go"
+    ]
+}

.vscode/launch.json

@@ -2,19 +2,33 @@
     "version": "0.2.0",
     "configurations": [
         {
-            "name": "Launch Go",
+            "name": "Python: Current File",
+            "type": "debugpy",
+            "request": "launch",
+            "program": "${file}",
+            "console": "integratedTerminal",
+            "justMyCode": false,
+            "cwd": "${fileDirname}",
+            "env": {
+                "OPENAI_API_BASE": "http://localhost:8080/v1",
+                "OPENAI_API_KEY": "abc"
+            }
+        },
+        {
+            "name": "Launch LocalAI API",
             "type": "go",
             "request": "launch",
             "mode": "debug",
-            "program": "${workspaceFolder}/main.go",
-            "args": [
-                "api"
-            ],
+            "program": "${workspaceRoot}",
+            "args": [],
             "env": {
-                "C_INCLUDE_PATH": "/workspace/go-llama:/workspace/go-gpt4all-j:/workspace/go-gpt2",
-                "LIBRARY_PATH": "/workspace/go-llama:/workspace/go-gpt4all-j:/workspace/go-gpt2",
-                "DEBUG": "true"
-            }
+                "LOCALAI_LOG_LEVEL": "debug",
+                "LOCALAI_P2P": "true",
+                "LOCALAI_FEDERATED": "true"
+            },
+            "buildFlags": ["-tags", "p2p tts", "-v"],
+            "envFile": "${workspaceFolder}/.env",
+            "cwd": "${workspaceRoot}"
         }
     ]
 }

.yamllint

@@ -0,0 +1,4 @@
+extends: default
+
+rules:
+    line-length: disable

CONTRIBUTING.md

@@ -0,0 +1,84 @@
+# Contributing to LocalAI
+
+Thank you for your interest in contributing to LocalAI! We appreciate your time and effort in helping to improve our project. Before you get started, please take a moment to review these guidelines.
+
+## Table of Contents
+
+- [Getting Started](#getting-started)
+  - [Prerequisites](#prerequisites)
+  - [Setting up the Development Environment](#setting-up-the-development-environment)
+- [Contributing](#contributing)
+  - [Submitting an Issue](#submitting-an-issue)
+  - [Creating a Pull Request (PR)](#creating-a-pull-request-pr)
+- [Coding Guidelines](#coding-guidelines)
+- [Testing](#testing)
+- [Documentation](#documentation)
+- [Community and Communication](#community-and-communication)
+
+## Getting Started
+
+### Prerequisites
+
+- Golang [1.21]
+- Git
+- macOS/Linux
+
+### Setting up the Development Environment and running localAI in the local environment
+
+1. Clone the repository: `git clone https://github.com/go-skynet/LocalAI.git`
+2. Navigate to the project directory: `cd LocalAI`
+3. Install the required dependencies ( see https://localai.io/basics/build/#build-localai-locally )
+4. Build LocalAI: `make build`
+5. Run LocalAI: `./local-ai`
+
+## Contributing
+
+We welcome contributions from everyone! To get started, follow these steps:
+
+### Submitting an Issue
+
+If you find a bug, have a feature request, or encounter any issues, please check the [issue tracker](https://github.com/go-skynet/LocalAI/issues) to see if a similar issue has already been reported. If not, feel free to [create a new issue](https://github.com/go-skynet/LocalAI/issues/new) and provide as much detail as possible.
+
+### Creating a Pull Request (PR)
+
+1. Fork the repository.
+2. Create a new branch with a descriptive name: `git checkout -b [branch name]`
+3. Make your changes and commit them.
+4. Push the changes to your fork: `git push origin [branch name]`
+5. Create a new pull request from your branch to the main project's `main` or `master` branch.
+6. Provide a clear description of your changes in the pull request.
+7. Make any requested changes during the review process.
+8. Once your PR is approved, it will be merged into the main project.
+
+## Coding Guidelines
+
+- No specific coding guidelines at the moment. Please make sure the code can be tested. The most popular lint tools like [`golangci-lint`](https://golangci-lint.run) can help you here.
+
+## Testing
+
+`make test` cannot handle all the model now. Please be sure to add a test case for the new features or the part was changed.
+
+### Running AIO tests
+
+All-In-One images has a set of tests that automatically verifies that most of the endpoints works correctly, a flow can be :
+
+```bash
+# Build the LocalAI docker image
+make DOCKER_IMAGE=local-ai docker
+
+# Build the corresponding AIO image
+BASE_IMAGE=local-ai DOCKER_AIO_IMAGE=local-ai-aio:test make docker-aio
+
+# Run the AIO e2e tests
+LOCALAI_IMAGE_TAG=test LOCALAI_IMAGE=local-ai-aio make run-e2e-aio
+```
+
+## Documentation
+
+We are welcome the contribution of the documents, please open new PR or create a new issue. The documentation is available under `docs/` https://github.com/mudler/LocalAI/tree/master/docs
+ 
+## Community and Communication
+
+- You can reach out via the Github issue tracker.
+- Open a new discussion at [Discussion](https://github.com/go-skynet/LocalAI/discussions)
+- Join the Discord channel [Discord](https://discord.gg/uJAeKSAGDy)

Dockerfile

@@ -1,10 +1,450 @@
-ARG GO_VERSION=1.20
-ARG BUILD_TYPE=
-FROM golang:$GO_VERSION
+ARG IMAGE_TYPE=extras
+ARG BASE_IMAGE=ubuntu:22.04
+ARG GRPC_BASE_IMAGE=${BASE_IMAGE}
+ARG INTEL_BASE_IMAGE=${BASE_IMAGE}
+
+# The requirements-core target is common to all images.  It should not be placed in requirements-core unless every single build will use it.
+FROM ${BASE_IMAGE} AS requirements-core
+
+USER root
+
+ARG GO_VERSION=1.22.6
+ARG CMAKE_VERSION=3.26.4
+ARG CMAKE_FROM_SOURCE=false
+ARG TARGETARCH
+ARG TARGETVARIANT
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV EXTERNAL_GRPC_BACKENDS="coqui:/build/backend/python/coqui/run.sh,transformers:/build/backend/python/transformers/run.sh,rerankers:/build/backend/python/rerankers/run.sh,bark:/build/backend/python/bark/run.sh,diffusers:/build/backend/python/diffusers/run.sh,faster-whisper:/build/backend/python/faster-whisper/run.sh,kokoro:/build/backend/python/kokoro/run.sh,vllm:/build/backend/python/vllm/run.sh,exllama2:/build/backend/python/exllama2/run.sh"
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        build-essential \
+        ccache \
+        ca-certificates \
+        curl libssl-dev \
+        git \
+        git-lfs \
+        unzip upx-ucl && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# Install CMake (the version in 22.04 is too old)
+RUN <<EOT bash
+    if [ "${CMAKE_FROM_SOURCE}}" = "true" ]; then
+        curl -L -s https://github.com/Kitware/CMake/releases/download/v${CMAKE_VERSION}/cmake-${CMAKE_VERSION}.tar.gz -o cmake.tar.gz && tar xvf cmake.tar.gz && cd cmake-${CMAKE_VERSION} && ./configure && make && make install
+    else
+        apt-get update && \
+        apt-get install -y \
+            cmake && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/*
+    fi
+EOT
+
+# Install Go
+RUN curl -L -s https://go.dev/dl/go${GO_VERSION}.linux-${TARGETARCH}.tar.gz | tar -C /usr/local -xz
+ENV PATH=$PATH:/root/go/bin:/usr/local/go/bin
+
+# Install grpc compilers
+RUN go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2 && \
+    go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+
+COPY --chmod=644 custom-ca-certs/* /usr/local/share/ca-certificates/
+RUN update-ca-certificates
+
+RUN test -n "$TARGETARCH" \
+    || (echo 'warn: missing $TARGETARCH, either set this `ARG` manually, or run using `docker buildkit`')
+
+# Use the variables in subsequent instructions
+RUN echo "Target Architecture: $TARGETARCH"
+RUN echo "Target Variant: $TARGETVARIANT"
+
+# Cuda
+ENV PATH=/usr/local/cuda/bin:${PATH}
+
+# HipBLAS requirements
+ENV PATH=/opt/rocm/bin:${PATH}
+
+# OpenBLAS requirements and stable diffusion
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        libopenblas-dev && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
 WORKDIR /build
-RUN apt-get update && apt-get install -y cmake libgomp1 libopenblas-dev libopenblas-base libopencv-dev libopencv-core-dev libopencv-core4.5 
+
+###################################
+###################################
+
+# The requirements-extras target is for any builds with IMAGE_TYPE=extras. It should not be placed in this target unless every IMAGE_TYPE=extras build will use it
+FROM requirements-core AS requirements-extras
+
+# Install uv as a system package
+RUN curl -LsSf https://astral.sh/uv/install.sh | UV_INSTALL_DIR=/usr/bin sh
+ENV PATH="/root/.cargo/bin:${PATH}"
+
+RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        espeak-ng \
+        espeak \
+        python3-pip \
+        python-is-python3 \
+        python3-dev llvm \
+        python3-venv && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/* && \
+    pip install --upgrade pip
+
+# Install grpcio-tools (the version in 22.04 is too old)
+RUN pip install --user grpcio-tools
+
+###################################
+###################################
+
+# The requirements-drivers target is for BUILD_TYPE specific items.  If you need to install something specific to CUDA, or specific to ROCM, it goes here.
+# This target will be built on top of requirements-core or requirements-extras as retermined by the IMAGE_TYPE build-arg
+FROM requirements-${IMAGE_TYPE} AS requirements-drivers
+
+ARG BUILD_TYPE
+ARG CUDA_MAJOR_VERSION=12
+ARG CUDA_MINOR_VERSION=0
+ARG SKIP_DRIVERS=false
+
+ENV BUILD_TYPE=${BUILD_TYPE}
+
+# Vulkan requirements
+RUN <<EOT bash
+    if [ "${BUILD_TYPE}" = "vulkan" ] && [ "${SKIP_DRIVERS}" = "false" ]; then
+        apt-get update && \
+        apt-get install -y  --no-install-recommends \
+            software-properties-common pciutils wget gpg-agent && \
+        wget -qO - https://packages.lunarg.com/lunarg-signing-key-pub.asc | apt-key add - && \
+        wget -qO /etc/apt/sources.list.d/lunarg-vulkan-jammy.list https://packages.lunarg.com/vulkan/lunarg-vulkan-jammy.list && \
+        apt-get update && \
+        apt-get install -y \
+            vulkan-sdk && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/*
+    fi
+EOT
+
+# CuBLAS requirements
+RUN <<EOT bash
+    if [ "${BUILD_TYPE}" = "cublas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then
+        apt-get update && \
+        apt-get install -y  --no-install-recommends \
+            software-properties-common pciutils
+        if [ "amd64" = "$TARGETARCH" ]; then
+            curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
+        fi
+        if [ "arm64" = "$TARGETARCH" ]; then
+            curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/arm64/cuda-keyring_1.1-1_all.deb
+        fi
+        dpkg -i cuda-keyring_1.1-1_all.deb && \
+        rm -f cuda-keyring_1.1-1_all.deb && \
+        apt-get update && \
+        apt-get install -y --no-install-recommends \
+            cuda-nvcc-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
+            libcufft-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
+            libcurand-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
+            libcublas-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
+            libcusparse-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
+            libcusolver-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/*
+    fi
+EOT
+
+# If we are building with clblas support, we need the libraries for the builds
+RUN if [ "${BUILD_TYPE}" = "clblas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then \
+        apt-get update && \
+        apt-get install -y --no-install-recommends \
+            libclblast-dev && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/* \
+    ; fi
+
+RUN if [ "${BUILD_TYPE}" = "hipblas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then \
+        apt-get update && \
+        apt-get install -y --no-install-recommends \
+            hipblas-dev \
+            rocblas-dev && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/* && \
+        # I have no idea why, but the ROCM lib packages don't trigger ldconfig after they install, which results in local-ai and others not being able
+        # to locate the libraries. We run ldconfig ourselves to work around this packaging deficiency
+        ldconfig \
+    ; fi
+
+###################################
+###################################
+
+# Temporary workaround for Intel's repository to work correctly
+# https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/APT-Repository-not-working-signatures-invalid/m-p/1599436/highlight/true#M36143
+# This is a temporary workaround until Intel fixes their repository
+FROM ${INTEL_BASE_IMAGE} AS intel
+RUN wget -qO - https://repositories.intel.com/gpu/intel-graphics.key | \
+gpg --yes --dearmor --output /usr/share/keyrings/intel-graphics.gpg
+RUN echo "deb [arch=amd64 signed-by=/usr/share/keyrings/intel-graphics.gpg] https://repositories.intel.com/gpu/ubuntu jammy/lts/2350 unified" > /etc/apt/sources.list.d/intel-graphics.list
+
+###################################
+###################################
+
+# The grpc target does one thing, it builds and installs GRPC.  This is in it's own layer so that it can be effectively cached by CI.
+# You probably don't need to change anything here, and if you do, make sure that CI is adjusted so that the cache continues to work.
+FROM ${GRPC_BASE_IMAGE} AS grpc
+
+# This is a bit of a hack, but it's required in order to be able to effectively cache this layer in CI
+ARG GRPC_MAKEFLAGS="-j4 -Otarget"
+ARG GRPC_VERSION=v1.65.0
+ARG CMAKE_FROM_SOURCE=false
+ARG CMAKE_VERSION=3.26.4
+
+ENV MAKEFLAGS=${GRPC_MAKEFLAGS}
+
+WORKDIR /build
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        ca-certificates \
+        build-essential curl libssl-dev \
+        git && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# Install CMake (the version in 22.04 is too old)
+RUN <<EOT bash
+    if [ "${CMAKE_FROM_SOURCE}}" = "true" ]; then
+        curl -L -s https://github.com/Kitware/CMake/releases/download/v${CMAKE_VERSION}/cmake-${CMAKE_VERSION}.tar.gz -o cmake.tar.gz && tar xvf cmake.tar.gz && cd cmake-${CMAKE_VERSION} && ./configure && make && make install
+    else
+        apt-get update && \
+        apt-get install -y \
+            cmake && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/*
+    fi
+EOT
+
+# We install GRPC to a different prefix here so that we can copy in only the build artifacts later
+# saves several hundred MB on the final docker image size vs copying in the entire GRPC source tree
+# and running make install in the target container
+RUN git clone --recurse-submodules --jobs 4 -b ${GRPC_VERSION} --depth 1 --shallow-submodules https://github.com/grpc/grpc && \
+    mkdir -p /build/grpc/cmake/build && \
+    cd /build/grpc/cmake/build && \
+    sed -i "216i\  TESTONLY" "../../third_party/abseil-cpp/absl/container/CMakeLists.txt" && \
+    cmake -DgRPC_INSTALL=ON -DgRPC_BUILD_TESTS=OFF -DCMAKE_INSTALL_PREFIX:PATH=/opt/grpc ../.. && \
+    make && \
+    make install && \
+    rm -rf /build
+
+###################################
+###################################
+
+# The builder-base target has the arguments, variables, and copies shared between full builder images and the uncompiled devcontainer
+
+FROM requirements-drivers AS builder-base
+
+ARG GO_TAGS="tts p2p"
+ARG GRPC_BACKENDS
+ARG MAKEFLAGS
+ARG LD_FLAGS="-s -w"
+
+ENV GRPC_BACKENDS=${GRPC_BACKENDS}
+ENV GO_TAGS=${GO_TAGS}
+ENV MAKEFLAGS=${MAKEFLAGS}
+ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
+ENV NVIDIA_REQUIRE_CUDA="cuda>=${CUDA_MAJOR_VERSION}.0"
+ENV NVIDIA_VISIBLE_DEVICES=all
+ENV LD_FLAGS=${LD_FLAGS}
+
+RUN echo "GO_TAGS: $GO_TAGS" && echo "TARGETARCH: $TARGETARCH"
+
+WORKDIR /build
+
+
+# We need protoc installed, and the version in 22.04 is too old.  We will create one as part installing the GRPC build below
+# but that will also being in a newer version of absl which stablediffusion cannot compile with.  This version of protoc is only
+# here so that we can generate the grpc code for the stablediffusion build
+RUN <<EOT bash
+    if [ "amd64" = "$TARGETARCH" ]; then
+        curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v27.1/protoc-27.1-linux-x86_64.zip -o protoc.zip && \
+        unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
+        rm protoc.zip
+    fi
+    if [ "arm64" = "$TARGETARCH" ]; then
+        curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v27.1/protoc-27.1-linux-aarch_64.zip -o protoc.zip && \
+        unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
+        rm protoc.zip
+    fi
+EOT
+
+###################################
+###################################
+
+# The builder target compiles LocalAI. This target is not the target that will be uploaded to the registry.
+# Adjustments to the build process should likely be made here.
+FROM builder-base AS builder
+
+# Install the pre-built GRPC
+COPY --from=grpc /opt/grpc /usr/local
+
+# Rebuild with defaults backends
+WORKDIR /build
+
 COPY . .
-RUN ln -s /usr/include/opencv4/opencv2/ /usr/include/opencv2
+COPY .git .
+
+RUN make prepare
+
+## Build the binary
+## If it's CUDA or hipblas, we want to skip some of the llama-compat backends to save space
+## We only leave the most CPU-optimized variant and the fallback for the cublas/hipblas build
+## (both will use CUDA or hipblas for the actual computation)
+RUN if [ "${BUILD_TYPE}" = "cublas" ] || [ "${BUILD_TYPE}" = "hipblas" ]; then \
+        SKIP_GRPC_BACKEND="backend-assets/grpc/llama-cpp-avx512 backend-assets/grpc/llama-cpp-avx backend-assets/grpc/llama-cpp-avx2" make build; \
+    else \
+        make build; \
+    fi
+
+RUN if [ ! -d "/build/sources/go-piper/piper-phonemize/pi/lib/" ]; then \
+        mkdir -p /build/sources/go-piper/piper-phonemize/pi/lib/ \
+        touch /build/sources/go-piper/piper-phonemize/pi/lib/keep \
+    ; fi
+
+###################################
+###################################
+
+# The devcontainer target is not used on CI. It is a target for developers to use locally -
+# rather than copying files it mounts them locally and leaves building to the developer
+
+FROM builder-base AS devcontainer
+
+ARG FFMPEG
+
+COPY --from=grpc /opt/grpc /usr/local
+
+COPY .devcontainer-scripts /.devcontainer-scripts
+
+# Add FFmpeg
+RUN if [ "${FFMPEG}" = "true" ]; then \
+        apt-get update && \
+        apt-get install -y --no-install-recommends \
+            ffmpeg && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/* \
+    ; fi
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        ssh less wget
+# For the devcontainer, leave apt functional in case additional devtools are needed at runtime.
+
+RUN go install github.com/go-delve/delve/cmd/dlv@latest
+
+RUN go install github.com/mikefarah/yq/v4@latest
+
+###################################
+###################################
+
+# This is the final target. The result of this target will be the image uploaded to the registry.
+# If you cannot find a more suitable place for an addition, this layer is a suitable place for it.
+FROM requirements-drivers
+
+ARG FFMPEG
+ARG BUILD_TYPE
+ARG TARGETARCH
+ARG IMAGE_TYPE=extras
+ARG EXTRA_BACKENDS
+ARG MAKEFLAGS
+
+ENV BUILD_TYPE=${BUILD_TYPE}
+ENV REBUILD=false
+ENV HEALTHCHECK_ENDPOINT=http://localhost:8080/readyz
+ENV MAKEFLAGS=${MAKEFLAGS}
+
+ARG CUDA_MAJOR_VERSION=12
+ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
+ENV NVIDIA_REQUIRE_CUDA="cuda>=${CUDA_MAJOR_VERSION}.0"
+ENV NVIDIA_VISIBLE_DEVICES=all
+
+# Add FFmpeg
+RUN if [ "${FFMPEG}" = "true" ]; then \
+        apt-get update && \
+        apt-get install -y --no-install-recommends \
+            ffmpeg && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/* \
+    ; fi
+
+WORKDIR /build
+
+# we start fresh & re-copy all assets because `make build` does not clean up nicely after itself
+# so when `entrypoint.sh` runs `make build` again (which it does by default), the build would fail
+# see https://github.com/go-skynet/LocalAI/pull/658#discussion_r1241971626 and
+# https://github.com/go-skynet/LocalAI/pull/434
+COPY . .
+
+COPY --from=builder /build/sources ./sources/
+COPY --from=grpc /opt/grpc /usr/local
+
 RUN make prepare-sources
+
+# Copy the binary
+COPY --from=builder /build/local-ai ./
+
+# Copy shared libraries for piper
+COPY --from=builder /build/sources/go-piper/piper-phonemize/pi/lib/* /usr/lib/
+
+# Change the shell to bash so we can use [[ tests below
+SHELL ["/bin/bash", "-c"]
+# We try to strike a balance between individual layer size (as that affects total push time) and total image size
+# Splitting the backends into more groups with fewer items results in a larger image, but a smaller size for the largest layer
+# Splitting the backends into fewer groups with more items results in a smaller image, but a larger size for the largest layer
+
+RUN if [[ ( "${IMAGE_TYPE}" == "extras ")]]; then \
+        apt-get -qq -y install espeak-ng \
+    ; fi
+
+RUN if [[ ( "${EXTRA_BACKENDS}" =~ "coqui" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/coqui \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "faster-whisper" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/faster-whisper \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "diffusers" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/diffusers \
+    ; fi
+
+RUN if [[ ( "${EXTRA_BACKENDS}" =~ "kokoro" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/kokoro \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "exllama2" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/exllama2 \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "transformers" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/transformers \
+    ; fi
+
+RUN if [[ ( "${EXTRA_BACKENDS}" =~ "vllm" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/vllm \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "bark" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/bark \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "rerankers" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/rerankers \
+    ; fi
+
+# Make sure the models directory exists
+RUN mkdir -p /build/models
+
+# Define the health check command
+HEALTHCHECK --interval=1m --timeout=10m --retries=10 \
+  CMD curl -f ${HEALTHCHECK_ENDPOINT} || exit 1
+
+VOLUME /build/models
 EXPOSE 8080
 ENTRYPOINT [ "/build/entrypoint.sh" ]

Dockerfile.aio

@@ -0,0 +1,8 @@
+ARG BASE_IMAGE=ubuntu:22.04
+
+FROM ${BASE_IMAGE} 
+
+RUN apt-get update && apt-get install -y pciutils && apt-get clean
+
+COPY aio/ /aio
+ENTRYPOINT [ "/aio/entrypoint.sh" ]

Dockerfile.dev

@@ -1,15 +0,0 @@
-ARG GO_VERSION=1.20
-ARG DEBIAN_VERSION=11
-ARG BUILD_TYPE=
-
-FROM golang:$GO_VERSION as builder
-WORKDIR /build
-RUN apt-get update && apt-get install -y cmake libgomp1 libopenblas-dev libopenblas-base libopencv-dev libopencv-core-dev libopencv-core4.5 
-RUN ln -s /usr/include/opencv4/opencv2/ /usr/include/opencv2
-COPY . .
-RUN make build
-
-FROM debian:$DEBIAN_VERSION
-COPY --from=builder /build/local-ai /usr/bin/local-ai
-EXPOSE 8080
-ENTRYPOINT [ "/usr/bin/local-ai" ]

Entitlements.plist

@@ -0,0 +1,10 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+<dict>
+    <key>com.apple.security.network.client</key>
+    <true/>
+    <key>com.apple.security.network.server</key>
+    <true/>
+</dict>
+</plist>

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2023 go-skynet authors
+Copyright (c) 2023-2025 Ettore Di Giacinto (mudler@localai.io)
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

Makefile

@@ -3,219 +3,477 @@ GOTEST=$(GOCMD) test
 GOVET=$(GOCMD) vet
 BINARY_NAME=local-ai
 
-GOLLAMA_VERSION?=7f9ae4246088f0c08ed322acbae21d69cd2eb547
-GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
-GPT4ALL_VERSION?=a07237e54fcdfdb351913587052ac061a2a7bdff
-GOGPT2_VERSION?=7bff56f0224502c1c9ed6258d2a17e8084628827
-RWKV_REPO?=https://github.com/donomii/go-rwkv.cpp
-RWKV_VERSION?=07166da10cb2a9e8854395a4f210464dcea76e47
-WHISPER_CPP_VERSION?=95b02d76b04d18e4ce37ed8353a1f0797f1717ea
-BERT_VERSION?=cea1ed76a7f48ef386a8e369f6c82c48cdf2d551
-BLOOMZ_VERSION?=e9366e82abdfe70565644fbfae9651976714efd1
-BUILD_TYPE?=
-CGO_LDFLAGS?=
-CUDA_LIBPATH?=/usr/local/cuda/lib64/
-STABLEDIFFUSION_VERSION?=c0748eca3642d58bcf9521108bcee46959c647dc
+DETECT_LIBS?=true
 
+# llama.cpp versions
+CPPLLAMA_VERSION?=6408210082cc0a61b992b487be7e2ff2efbb9e36
+
+# whisper.cpp version
+WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp
+WHISPER_CPP_VERSION?=6266a9f9e56a5b925e9892acf650f3eb1245814d
+
+# go-piper version
+PIPER_REPO?=https://github.com/mudler/go-piper
+PIPER_VERSION?=e10ca041a885d4a8f3871d52924b47792d5e5aa0
+
+# bark.cpp
+BARKCPP_REPO?=https://github.com/PABannier/bark.cpp.git
+BARKCPP_VERSION?=v1.0.0
+
+# stablediffusion.cpp (ggml)
+STABLEDIFFUSION_GGML_REPO?=https://github.com/richiejp/stable-diffusion.cpp
+STABLEDIFFUSION_GGML_VERSION?=53e3b17eb3d0b5760ced06a1f98320b68b34aaae
+
+ONNX_VERSION?=1.20.0
+ONNX_ARCH?=x64
+ONNX_OS?=linux
+
+export BUILD_TYPE?=
+export STABLE_BUILD_TYPE?=$(BUILD_TYPE)
+export CMAKE_ARGS?=
+export BACKEND_LIBS?=
+
+CGO_LDFLAGS?=
+CGO_LDFLAGS_WHISPER?=
+CGO_LDFLAGS_WHISPER+=-lggml
+CUDA_LIBPATH?=/usr/local/cuda/lib64/
 GO_TAGS?=
+BUILD_ID?=
+NATIVE?=false
+
+TEST_DIR=/tmp/test
+
+TEST_FLAKES?=5
+
+RANDOM := $(shell bash -c 'echo $$RANDOM')
+
+VERSION?=$(shell git describe --always --tags || echo "dev" )
+# go tool nm ./local-ai | grep Commit
+LD_FLAGS?=-s -w
+override LD_FLAGS += -X "github.com/mudler/LocalAI/internal.Version=$(VERSION)"
+override LD_FLAGS += -X "github.com/mudler/LocalAI/internal.Commit=$(shell git rev-parse HEAD)"
 
 OPTIONAL_TARGETS?=
 
+export OS := $(shell uname -s)
+ARCH := $(shell uname -m)
 GREEN  := $(shell tput -Txterm setaf 2)
 YELLOW := $(shell tput -Txterm setaf 3)
 WHITE  := $(shell tput -Txterm setaf 7)
 CYAN   := $(shell tput -Txterm setaf 6)
 RESET  := $(shell tput -Txterm sgr0)
 
-C_INCLUDE_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-gpt2:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
-LIBRARY_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-gpt2:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
+UPX?=
+# check if upx exists
+ifeq (, $(shell which upx))
+	UPX=
+else
+	UPX=$(shell which upx)
+endif
+
+# Default Docker bridge IP
+E2E_BRIDGE_IP?=172.17.0.1
+
+ifndef UNAME_S
+UNAME_S := $(shell uname -s)
+endif
+
+# IF native is false, we add -DGGML_NATIVE=OFF to CMAKE_ARGS
+ifeq ($(NATIVE),false)
+	CMAKE_ARGS+=-DGGML_NATIVE=OFF
+endif
+
+# Detect if we are running on arm64
+ifneq (,$(findstring aarch64,$(shell uname -m)))
+	ONNX_ARCH=aarch64
+endif
+
+ifeq ($(OS),Darwin)
+	ONNX_OS=osx
+	ifneq (,$(findstring aarch64,$(shell uname -m)))
+		ONNX_ARCH=arm64
+	else ifneq (,$(findstring arm64,$(shell uname -m)))
+		ONNX_ARCH=arm64
+	else
+		ONNX_ARCH=x86_64
+	endif
+
+	ifeq ($(OSX_SIGNING_IDENTITY),)
+		OSX_SIGNING_IDENTITY := $(shell security find-identity -v -p codesigning | grep '"' | head -n 1 | sed -E 's/.*"(.*)"/\1/')
+	endif
+
+	# on OSX, if BUILD_TYPE is blank, we should default to use Metal
+	ifeq ($(BUILD_TYPE),)
+		BUILD_TYPE=metal
+	# disable metal if on Darwin and any other value is explicitly passed.
+	else ifneq ($(BUILD_TYPE),metal)
+		CMAKE_ARGS+=-DGGML_METAL=OFF
+		export GGML_NO_ACCELERATE=1
+		export GGML_NO_METAL=1
+	endif
+
+	ifeq ($(BUILD_TYPE),metal)
+#			-lcblas 	removed: it seems to always be listed as a duplicate flag.
+		CGO_LDFLAGS += -framework Accelerate
+	endif
+else
+CGO_LDFLAGS_WHISPER+=-lgomp
+endif
 
 ifeq ($(BUILD_TYPE),openblas)
 	CGO_LDFLAGS+=-lopenblas
+	export GGML_OPENBLAS=1
 endif
 
 ifeq ($(BUILD_TYPE),cublas)
 	CGO_LDFLAGS+=-lcublas -lcudart -L$(CUDA_LIBPATH)
+	export GGML_CUDA=1
+	CGO_LDFLAGS_WHISPER+=-L$(CUDA_LIBPATH)/stubs/ -lcuda -lcufft
 endif
 
-
-ifeq ($(GO_TAGS),stablediffusion)
-	OPTIONAL_TARGETS+=go-stable-diffusion/libstablediffusion.a
+ifeq ($(BUILD_TYPE),vulkan)
+	CMAKE_ARGS+=-DGGML_VULKAN=1
 endif
 
-.PHONY: all test build vendor
+ifneq (,$(findstring sycl,$(BUILD_TYPE)))
+	export GGML_SYCL=1
+endif
+
+ifeq ($(BUILD_TYPE),sycl_f16)
+	export GGML_SYCL_F16=1
+endif
+
+ifeq ($(BUILD_TYPE),hipblas)
+	ROCM_HOME ?= /opt/rocm
+	ROCM_PATH ?= /opt/rocm
+	LD_LIBRARY_PATH ?= /opt/rocm/lib:/opt/rocm/llvm/lib
+	export CXX=$(ROCM_HOME)/llvm/bin/clang++
+	export CC=$(ROCM_HOME)/llvm/bin/clang
+	export STABLE_BUILD_TYPE=
+	export GGML_HIP=1
+	GPU_TARGETS ?= gfx900,gfx906,gfx908,gfx940,gfx941,gfx942,gfx90a,gfx1030,gfx1031,gfx1100,gfx1101
+	AMDGPU_TARGETS ?= "$(GPU_TARGETS)"
+	CMAKE_ARGS+=-DGGML_HIP=ON -DAMDGPU_TARGETS="$(AMDGPU_TARGETS)" -DGPU_TARGETS="$(GPU_TARGETS)"
+	CGO_LDFLAGS += -O3 --rtlib=compiler-rt -unwindlib=libgcc -lhipblas -lrocblas --hip-link -L${ROCM_HOME}/lib/llvm/lib
+endif
+
+ifeq ($(BUILD_TYPE),metal)
+	CGO_LDFLAGS+=-framework Foundation -framework Metal -framework MetalKit -framework MetalPerformanceShaders
+	export GGML_METAL=1
+endif
+
+ifeq ($(BUILD_TYPE),clblas)
+	CGO_LDFLAGS+=-lOpenCL -lclblast
+	export GGML_OPENBLAS=1
+endif
+
+# glibc-static or glibc-devel-static required
+ifeq ($(STATIC),true)
+	LD_FLAGS+=-linkmode external -extldflags -static
+endif
+
+ifeq ($(findstring tts,$(GO_TAGS)),tts)
+#	OPTIONAL_TARGETS+=go-piper/libpiper_binding.a
+#	OPTIONAL_TARGETS+=backend-assets/espeak-ng-data
+	PIPER_CGO_CXXFLAGS+=-I$(CURDIR)/sources/go-piper/piper/src/cpp -I$(CURDIR)/sources/go-piper/piper/build/fi/include -I$(CURDIR)/sources/go-piper/piper/build/pi/include -I$(CURDIR)/sources/go-piper/piper/build/si/include
+	PIPER_CGO_LDFLAGS+=-L$(CURDIR)/sources/go-piper/piper/build/fi/lib -L$(CURDIR)/sources/go-piper/piper/build/pi/lib -L$(CURDIR)/sources/go-piper/piper/build/si/lib -lfmt -lspdlog -lucd
+	OPTIONAL_GRPC+=backend-assets/grpc/piper
+endif
+
+ALL_GRPC_BACKENDS=backend-assets/grpc/huggingface
+ALL_GRPC_BACKENDS+=backend-assets/grpc/llama-cpp-avx
+ALL_GRPC_BACKENDS+=backend-assets/grpc/llama-cpp-avx2
+ALL_GRPC_BACKENDS+=backend-assets/grpc/llama-cpp-avx512
+ALL_GRPC_BACKENDS+=backend-assets/grpc/llama-cpp-fallback
+ALL_GRPC_BACKENDS+=backend-assets/grpc/llama-cpp-grpc
+ALL_GRPC_BACKENDS+=backend-assets/util/llama-cpp-rpc-server
+ALL_GRPC_BACKENDS+=backend-assets/grpc/whisper
+
+ifeq ($(ONNX_OS),linux)
+ifeq ($(ONNX_ARCH),x64)
+	ALL_GRPC_BACKENDS+=backend-assets/grpc/bark-cpp
+	ALL_GRPC_BACKENDS+=backend-assets/grpc/stablediffusion-ggml
+endif
+endif
+
+ALL_GRPC_BACKENDS+=backend-assets/grpc/local-store
+ALL_GRPC_BACKENDS+=backend-assets/grpc/silero-vad
+ALL_GRPC_BACKENDS+=$(OPTIONAL_GRPC)
+# Use filter-out to remove the specified backends
+ALL_GRPC_BACKENDS := $(filter-out $(SKIP_GRPC_BACKEND),$(ALL_GRPC_BACKENDS))
+
+GRPC_BACKENDS?=$(ALL_GRPC_BACKENDS) $(OPTIONAL_GRPC)
+TEST_PATHS?=./api/... ./pkg/... ./core/...
+
+# If empty, then we build all
+ifeq ($(GRPC_BACKENDS),)
+	GRPC_BACKENDS=$(ALL_GRPC_BACKENDS)
+endif
+
+ifeq ($(BUILD_API_ONLY),true)
+	GRPC_BACKENDS=
+endif
+
+.PHONY: all test build vendor get-sources prepare-sources prepare
 
 all: help
 
-## GPT4ALL
-gpt4all:
-	git clone --recurse-submodules $(GPT4ALL_REPO) gpt4all
-	cd gpt4all && git checkout -b build $(GPT4ALL_VERSION) && git submodule update --init --recursive --depth 1
-	# This is hackish, but needed as both go-llama and go-gpt4allj have their own version of ggml..
-	@find ./gpt4all -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_gptj_/g' {} +
-	@find ./gpt4all -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_gptj_/g' {} +
-	@find ./gpt4all -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_gptj_/g' {} +
-	@find ./gpt4all -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_/gptj_/g' {} +
-	@find ./gpt4all -type f -name "*.h" -exec sed -i'' -e 's/gpt_/gptj_/g' {} +
-	@find ./gpt4all -type f -name "*.h" -exec sed -i'' -e 's/set_console_color/set_gptj_console_color/g' {} +
-	@find ./gpt4all -type f -name "*.cpp" -exec sed -i'' -e 's/set_console_color/set_gptj_console_color/g' {} +
-	@find ./gpt4all -type f -name "*.cpp" -exec sed -i'' -e 's/llama_/gptjllama_/g' {} +
-	@find ./gpt4all -type f -name "*.go" -exec sed -i'' -e 's/llama_/gptjllama_/g' {} +
-	@find ./gpt4all -type f -name "*.h" -exec sed -i'' -e 's/llama_/gptjllama_/g' {} +
-	@find ./gpt4all -type f -name "*.txt" -exec sed -i'' -e 's/llama_/gptjllama_/g' {} +
-	@find ./gpt4all -type f -name "*.cpp" -exec sed -i'' -e 's/json_/json_gptj_/g' {} +
-	@find ./gpt4all -type f -name "*.cpp" -exec sed -i'' -e 's/void replace/void json_gptj_replace/g' {} +
-	@find ./gpt4all -type f -name "*.cpp" -exec sed -i'' -e 's/::replace/::json_gptj_replace/g' {} +
-	mv ./gpt4all/gpt4all-backend/llama.cpp/llama_util.h ./gpt4all/gpt4all-backend/llama.cpp/gptjllama_util.h
+## bark.cpp
+sources/bark.cpp:
+	git clone --recursive $(BARKCPP_REPO) sources/bark.cpp && \
+	cd sources/bark.cpp && \
+	git checkout $(BARKCPP_VERSION) && \
+	git submodule update --init --recursive --depth 1 --single-branch
 
-## BERT embeddings
-go-bert:
-	git clone --recurse-submodules https://github.com/go-skynet/go-bert.cpp go-bert
-	cd go-bert && git checkout -b build $(BERT_VERSION) && git submodule update --init --recursive --depth 1
-	@find ./go-bert -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_bert_/g' {} +
-	@find ./go-bert -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_bert_/g' {} +
-	@find ./go-bert -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_bert_/g' {} +
+sources/bark.cpp/build/libbark.a: sources/bark.cpp
+	cd sources/bark.cpp && \
+	mkdir -p build && \
+	cd build && \
+	cmake $(CMAKE_ARGS) .. && \
+	cmake --build . --config Release
 
-## stable diffusion
-go-stable-diffusion:
-	git clone --recurse-submodules https://github.com/mudler/go-stable-diffusion go-stable-diffusion
-	cd go-stable-diffusion && git checkout -b build $(STABLEDIFFUSION_VERSION) && git submodule update --init --recursive --depth 1
+backend/go/bark/libbark.a: sources/bark.cpp/build/libbark.a
+	$(MAKE) -C backend/go/bark libbark.a
 
-go-stable-diffusion/libstablediffusion.a:
-	$(MAKE) -C go-stable-diffusion libstablediffusion.a
+## go-piper
+sources/go-piper:
+	mkdir -p sources/go-piper
+	cd sources/go-piper && \
+	git init && \
+	git remote add origin $(PIPER_REPO) && \
+	git fetch origin && \
+	git checkout $(PIPER_VERSION) && \
+	git submodule update --init --recursive --depth 1 --single-branch
 
-## RWKV
-go-rwkv:
-	git clone --recurse-submodules $(RWKV_REPO) go-rwkv
-	cd go-rwkv && git checkout -b build $(RWKV_VERSION) && git submodule update --init --recursive --depth 1
-	@find ./go-rwkv -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_rwkv_/g' {} +
-	@find ./go-rwkv -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_rwkv_/g' {} +
-	@find ./go-rwkv -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_rwkv_/g' {} +
+sources/go-piper/libpiper_binding.a: sources/go-piper
+	$(MAKE) -C sources/go-piper libpiper_binding.a example/main piper.o
 
-go-rwkv/librwkv.a: go-rwkv
-	cd go-rwkv && cd rwkv.cpp &&	cmake . -DRWKV_BUILD_SHARED_LIBRARY=OFF &&	cmake --build . && 	cp librwkv.a .. && cp ggml/src/libggml.a ..
+## stablediffusion (ggml)
+sources/stablediffusion-ggml.cpp:
+	git clone --recursive $(STABLEDIFFUSION_GGML_REPO) sources/stablediffusion-ggml.cpp && \
+	cd sources/stablediffusion-ggml.cpp && \
+	git checkout $(STABLEDIFFUSION_GGML_VERSION) && \
+	git submodule update --init --recursive --depth 1 --single-branch
 
-## bloomz
-bloomz:
-	git clone --recurse-submodules https://github.com/go-skynet/bloomz.cpp bloomz
-	@find ./bloomz -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_bloomz_/g' {} +
-	@find ./bloomz -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_bloomz_/g' {} +
-	@find ./bloomz -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_bloomz_/g' {} +
-	@find ./bloomz -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_/gpt_bloomz_/g' {} +
-	@find ./bloomz -type f -name "*.h" -exec sed -i'' -e 's/gpt_/gpt_bloomz_/g' {} +
+backend/go/image/stablediffusion-ggml/libsd.a: sources/stablediffusion-ggml.cpp
+	$(MAKE) -C backend/go/image/stablediffusion-ggml build/libstable-diffusion.a
+	$(MAKE) -C backend/go/image/stablediffusion-ggml libsd.a
 
-bloomz/libbloomz.a: bloomz
-	cd bloomz && make libbloomz.a
+backend-assets/grpc/stablediffusion-ggml: backend/go/image/stablediffusion-ggml/libsd.a backend-assets/grpc
+	$(MAKE) -C backend/go/image/stablediffusion-ggml CGO_LDFLAGS="$(CGO_LDFLAGS)" stablediffusion-ggml
 
-go-bert/libgobert.a: go-bert
-	$(MAKE) -C go-bert libgobert.a
+sources/onnxruntime:
+	mkdir -p sources/onnxruntime
+	curl -L https://github.com/microsoft/onnxruntime/releases/download/v$(ONNX_VERSION)/onnxruntime-$(ONNX_OS)-$(ONNX_ARCH)-$(ONNX_VERSION).tgz -o sources/onnxruntime/onnxruntime-$(ONNX_OS)-$(ONNX_ARCH)-$(ONNX_VERSION).tgz
+	cd sources/onnxruntime && tar -xvf onnxruntime-$(ONNX_OS)-$(ONNX_ARCH)-$(ONNX_VERSION).tgz && rm onnxruntime-$(ONNX_OS)-$(ONNX_ARCH)-$(ONNX_VERSION).tgz
+	cd sources/onnxruntime && mv onnxruntime-$(ONNX_OS)-$(ONNX_ARCH)-$(ONNX_VERSION)/* ./
 
-gpt4all/gpt4all-bindings/golang/libgpt4all.a: gpt4all
-	$(MAKE) -C gpt4all/gpt4all-bindings/golang/ libgpt4all.a
+backend-assets/lib/libonnxruntime.so.1: backend-assets/lib sources/onnxruntime
+	cp -rfv sources/onnxruntime/lib/* backend-assets/lib/
+ifeq ($(OS),Darwin)
+	mv backend-assets/lib/libonnxruntime.$(ONNX_VERSION).dylib backend-assets/lib/libonnxruntime.dylib
+else
+	mv backend-assets/lib/libonnxruntime.so.$(ONNX_VERSION) backend-assets/lib/libonnxruntime.so.1
+endif
 
-## CEREBRAS GPT
-go-gpt2: 
-	git clone --recurse-submodules https://github.com/go-skynet/go-gpt2.cpp go-gpt2
-	cd go-gpt2 && git checkout -b build $(GOGPT2_VERSION) && git submodule update --init --recursive --depth 1
-	# This is hackish, but needed as both go-llama and go-gpt4allj have their own version of ggml..
-	@find ./go-gpt2 -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/json_/json_gpt2_/g' {} +
+## whisper
+sources/whisper.cpp:
+	mkdir -p sources/whisper.cpp
+	cd sources/whisper.cpp && \
+	git init && \
+	git remote add origin $(WHISPER_REPO) && \
+	git fetch origin && \
+	git checkout $(WHISPER_CPP_VERSION) && \
+	git submodule update --init --recursive --depth 1 --single-branch
 
-go-gpt2/libgpt2.a: go-gpt2
-	$(MAKE) -C go-gpt2 libgpt2.a
+sources/whisper.cpp/libwhisper.a: sources/whisper.cpp
+	cd sources/whisper.cpp && $(MAKE) libwhisper.a libggml.a
 
-whisper.cpp:
-	git clone https://github.com/ggerganov/whisper.cpp.git
-	cd whisper.cpp && git checkout -b build $(WHISPER_CPP_VERSION) && git submodule update --init --recursive --depth 1
-	@find ./whisper.cpp -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_whisper_/g' {} +
-	@find ./whisper.cpp -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_whisper_/g' {} +
-	@find ./whisper.cpp -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_whisper_/g' {} +
-
-whisper.cpp/libwhisper.a: whisper.cpp
-	cd whisper.cpp && make libwhisper.a
-
-go-llama:
-	git clone --recurse-submodules https://github.com/go-skynet/go-llama.cpp go-llama
-	cd go-llama && git checkout -b build $(GOLLAMA_VERSION) && git submodule update --init --recursive --depth 1
-
-go-llama/libbinding.a: go-llama 
-	$(MAKE) -C go-llama BUILD_TYPE=$(BUILD_TYPE) libbinding.a
+get-sources: sources/go-piper sources/stablediffusion-ggml.cpp sources/bark.cpp sources/whisper.cpp backend/cpp/llama/llama.cpp
 
 replace:
-	$(GOCMD) mod edit -replace github.com/go-skynet/go-llama.cpp=$(shell pwd)/go-llama
-	$(GOCMD) mod edit -replace github.com/nomic-ai/gpt4all/gpt4all-bindings/golang=$(shell pwd)/gpt4all/gpt4all-bindings/golang
-	$(GOCMD) mod edit -replace github.com/go-skynet/go-gpt2.cpp=$(shell pwd)/go-gpt2
-	$(GOCMD) mod edit -replace github.com/donomii/go-rwkv.cpp=$(shell pwd)/go-rwkv
-	$(GOCMD) mod edit -replace github.com/ggerganov/whisper.cpp=$(shell pwd)/whisper.cpp
-	$(GOCMD) mod edit -replace github.com/go-skynet/go-bert.cpp=$(shell pwd)/go-bert
-	$(GOCMD) mod edit -replace github.com/go-skynet/bloomz.cpp=$(shell pwd)/bloomz
-	$(GOCMD) mod edit -replace github.com/mudler/go-stable-diffusion=$(shell pwd)/go-stable-diffusion
+	$(GOCMD) mod edit -replace github.com/ggerganov/whisper.cpp=$(CURDIR)/sources/whisper.cpp
+	$(GOCMD) mod edit -replace github.com/ggerganov/whisper.cpp/bindings/go=$(CURDIR)/sources/whisper.cpp/bindings/go
+	$(GOCMD) mod edit -replace github.com/mudler/go-piper=$(CURDIR)/sources/go-piper
 
-prepare-sources: go-llama go-gpt2 gpt4all go-rwkv whisper.cpp go-bert bloomz go-stable-diffusion replace
+dropreplace:
+	$(GOCMD) mod edit -dropreplace github.com/ggerganov/whisper.cpp
+	$(GOCMD) mod edit -dropreplace github.com/ggerganov/whisper.cpp/bindings/go
+	$(GOCMD) mod edit -dropreplace github.com/mudler/go-piper
+
+prepare-sources: get-sources replace
 	$(GOCMD) mod download
 
 ## GENERIC
 rebuild: ## Rebuilds the project
-	$(MAKE) -C go-llama clean
-	$(MAKE) -C gpt4all/gpt4all-bindings/golang/ clean
-	$(MAKE) -C go-gpt2 clean
-	$(MAKE) -C go-rwkv clean
-	$(MAKE) -C whisper.cpp clean
-	$(MAKE) -C go-stable-diffusion clean
-	$(MAKE) -C go-bert clean
-	$(MAKE) -C bloomz clean
+	$(GOCMD) clean -cache
+	$(MAKE) -C sources/whisper.cpp clean
+	$(MAKE) -C sources/go-piper clean
 	$(MAKE) build
 
-prepare: prepare-sources gpt4all/gpt4all-bindings/golang/libgpt4all.a $(OPTIONAL_TARGETS) go-llama/libbinding.a go-bert/libgobert.a go-gpt2/libgpt2.a go-rwkv/librwkv.a whisper.cpp/libwhisper.a bloomz/libbloomz.a  ## Prepares for building
+prepare: prepare-sources $(OPTIONAL_TARGETS)
 
 clean: ## Remove build related file
-	rm -fr ./go-llama
-	rm -rf ./gpt4all
-	rm -rf ./go-stable-diffusion
-	rm -rf ./go-gpt2
-	rm -rf ./go-rwkv
-	rm -rf ./go-bert
-	rm -rf ./bloomz
-	rm -rf ./whisper.cpp
+	$(GOCMD) clean -cache
+	rm -f prepare
+	rm -rf ./sources
 	rm -rf $(BINARY_NAME)
+	rm -rf release/
+	rm -rf backend-assets/*
+	$(MAKE) -C backend/cpp/grpc clean
+	$(MAKE) -C backend/go/bark clean
+	$(MAKE) -C backend/cpp/llama clean
+	$(MAKE) -C backend/go/image/stablediffusion-ggml clean
+	rm -rf backend/cpp/llama-* || true
+	$(MAKE) dropreplace
+	$(MAKE) protogen-clean
+	rmdir pkg/grpc/proto || true
+
+clean-tests:
+	rm -rf test-models
+	rm -rf test-dir
+	rm -rf core/http/backend-assets
+
+clean-dc: clean
+	cp -r /build/backend-assets /workspace/backend-assets
 
 ## Build:
-
-build: prepare ## Build the project
+build: prepare backend-assets grpcs ## Build the project
 	$(info ${GREEN}I local-ai build info:${RESET})
 	$(info ${GREEN}I BUILD_TYPE: ${YELLOW}$(BUILD_TYPE)${RESET})
 	$(info ${GREEN}I GO_TAGS: ${YELLOW}$(GO_TAGS)${RESET})
-	CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=${C_INCLUDE_PATH} LIBRARY_PATH=${LIBRARY_PATH} $(GOCMD) build -tags "$(GO_TAGS)" -x -o $(BINARY_NAME) ./
+	$(info ${GREEN}I LD_FLAGS: ${YELLOW}$(LD_FLAGS)${RESET})
+	$(info ${GREEN}I UPX: ${YELLOW}$(UPX)${RESET})
+ifneq ($(BACKEND_LIBS),)
+	$(MAKE) backend-assets/lib
+	cp -f $(BACKEND_LIBS) backend-assets/lib/
+endif
+	CGO_LDFLAGS="$(CGO_LDFLAGS)" $(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o $(BINARY_NAME) ./
 
-generic-build: ## Build the project using generic
-	BUILD_TYPE="generic" $(MAKE) build
+build-minimal:
+	BUILD_GRPC_FOR_BACKEND_LLAMA=true GRPC_BACKENDS="backend-assets/grpc/llama-cpp-avx2" GO_TAGS=p2p $(MAKE) build
+
+build-api:
+	BUILD_GRPC_FOR_BACKEND_LLAMA=true BUILD_API_ONLY=true GO_TAGS=p2p $(MAKE) build
+
+backend-assets/lib:
+	mkdir -p backend-assets/lib
+
+dist:
+	$(MAKE) backend-assets/grpc/llama-cpp-avx2
+ifeq ($(DETECT_LIBS),true)
+	scripts/prepare-libs.sh backend-assets/grpc/llama-cpp-avx2
+endif
+ifeq ($(OS),Darwin)
+	BUILD_TYPE=none $(MAKE) backend-assets/grpc/llama-cpp-fallback
+else
+	$(MAKE) backend-assets/grpc/llama-cpp-cuda
+	$(MAKE) backend-assets/grpc/llama-cpp-hipblas
+	$(MAKE) backend-assets/grpc/llama-cpp-sycl_f16
+	$(MAKE) backend-assets/grpc/llama-cpp-sycl_f32
+endif
+	GO_TAGS="tts p2p" $(MAKE) build
+ifeq ($(DETECT_LIBS),true)
+	scripts/prepare-libs.sh backend-assets/grpc/piper
+endif
+	GO_TAGS="tts p2p" STATIC=true $(MAKE) build
+	mkdir -p release
+# if BUILD_ID is empty, then we don't append it to the binary name
+ifeq ($(BUILD_ID),)
+	cp $(BINARY_NAME) release/$(BINARY_NAME)-$(OS)-$(ARCH)
+	shasum -a 256 release/$(BINARY_NAME)-$(OS)-$(ARCH) > release/$(BINARY_NAME)-$(OS)-$(ARCH).sha256
+else
+	cp $(BINARY_NAME) release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-$(ARCH)
+	shasum -a 256 release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-$(ARCH) > release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-$(ARCH).sha256
+endif
+
+dist-cross-linux-arm64:
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_NATIVE=off" GRPC_BACKENDS="backend-assets/grpc/llama-cpp-fallback backend-assets/grpc/llama-cpp-grpc backend-assets/util/llama-cpp-rpc-server" GO_TAGS="p2p" \
+	STATIC=true $(MAKE) build
+	mkdir -p release
+# if BUILD_ID is empty, then we don't append it to the binary name
+ifeq ($(BUILD_ID),)
+	cp $(BINARY_NAME) release/$(BINARY_NAME)-$(OS)-arm64
+	shasum -a 256 release/$(BINARY_NAME)-$(OS)-arm64 > release/$(BINARY_NAME)-$(OS)-arm64.sha256
+else
+	cp $(BINARY_NAME) release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-arm64
+	shasum -a 256 release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-arm64 > release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-arm64.sha256
+endif
+
+osx-signed: build
+	codesign --deep --force --sign "$(OSX_SIGNING_IDENTITY)" --entitlements "./Entitlements.plist" "./$(BINARY_NAME)"
 
 ## Run
 run: prepare ## run local-ai
-	CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=${C_INCLUDE_PATH} LIBRARY_PATH=${LIBRARY_PATH} $(GOCMD) run ./main.go
+	CGO_LDFLAGS="$(CGO_LDFLAGS)" $(GOCMD) run ./
 
-test-models/testmodel:
+test-models/testmodel.ggml:
 	mkdir test-models
 	mkdir test-dir
-	wget https://huggingface.co/concedo/cerebras-111M-ggml/resolve/main/cerberas-111m-q4_0.bin -O test-models/testmodel
-	wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin -O test-models/whisper-en
-	wget https://huggingface.co/skeskinen/ggml/resolve/main/all-MiniLM-L6-v2/ggml-model-q4_0.bin -O test-models/bert
-	wget https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav -O test-dir/audio.wav
-	wget https://huggingface.co/imxcstar/rwkv-4-raven-ggml/resolve/main/RWKV-4-Raven-1B5-v11-Eng99%25-Other1%25-20230425-ctx4096-16_Q4_2.bin -O test-models/rwkv
-	wget https://raw.githubusercontent.com/saharNooby/rwkv.cpp/5eb8f09c146ea8124633ab041d9ea0b1f1db4459/rwkv/20B_tokenizer.json -O test-models/rwkv.tokenizer.json
-	cp tests/fixtures/* test-models
+	wget -q https://huggingface.co/RichardErkhov/Qwen_-_Qwen2-1.5B-Instruct-gguf/resolve/main/Qwen2-1.5B-Instruct.Q2_K.gguf -O test-models/testmodel.ggml
+	wget -q https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin -O test-models/whisper-en
+	wget -q https://huggingface.co/mudler/all-MiniLM-L6-v2/resolve/main/ggml-model-q4_0.bin -O test-models/bert
+	wget -q https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav -O test-dir/audio.wav
+	cp tests/models_fixtures/* test-models
 
-test: prepare test-models/testmodel
-	cp tests/fixtures/* test-models
-	@C_INCLUDE_PATH=${C_INCLUDE_PATH} LIBRARY_PATH=${LIBRARY_PATH} TEST_DIR=$(abspath ./)/test-dir/ CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models $(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo -v -r ./api
+prepare-test: grpcs
+	cp -rf backend-assets core/http
+	cp tests/models_fixtures/* test-models
+
+test: prepare test-models/testmodel.ggml grpcs
+	@echo 'Running tests'
+	export GO_TAGS="tts debug"
+	$(MAKE) prepare-test
+	HUGGINGFACE_GRPC=$(abspath ./)/backend/python/transformers/run.sh TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="!llama-gguf"  --flake-attempts $(TEST_FLAKES) --fail-fast -v -r $(TEST_PATHS)
+	$(MAKE) test-llama-gguf
+	$(MAKE) test-tts
+	$(MAKE) test-stablediffusion
+
+prepare-e2e:
+	mkdir -p $(TEST_DIR)
+	cp -rfv $(abspath ./tests/e2e-fixtures)/gpu.yaml $(TEST_DIR)/gpu.yaml
+	test -e $(TEST_DIR)/ggllm-test-model.bin || wget -q https://huggingface.co/TheBloke/CodeLlama-7B-Instruct-GGUF/resolve/main/codellama-7b-instruct.Q2_K.gguf -O $(TEST_DIR)/ggllm-test-model.bin
+	docker build --build-arg GRPC_BACKENDS="$(GRPC_BACKENDS)" --build-arg IMAGE_TYPE=core --build-arg BUILD_TYPE=$(BUILD_TYPE) --build-arg CUDA_MAJOR_VERSION=12 --build-arg CUDA_MINOR_VERSION=0 --build-arg FFMPEG=true -t localai-tests .
+
+run-e2e-image:
+	ls -liah $(abspath ./tests/e2e-fixtures)
+	docker run -p 5390:8080 -e MODELS_PATH=/models -e THREADS=1 -e DEBUG=true -d --rm -v $(TEST_DIR):/models --gpus all --name e2e-tests-$(RANDOM) localai-tests
+
+run-e2e-aio: protogen-go
+	@echo 'Running e2e AIO tests'
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --flake-attempts $(TEST_FLAKES) -v -r ./tests/e2e-aio
+
+test-e2e:
+	@echo 'Running e2e tests'
+	BUILD_TYPE=$(BUILD_TYPE) \
+	LOCALAI_API=http://$(E2E_BRIDGE_IP):5390/v1 \
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --flake-attempts $(TEST_FLAKES) -v -r ./tests/e2e
+
+teardown-e2e:
+	rm -rf $(TEST_DIR) || true
+	docker stop $$(docker ps -q --filter ancestor=localai-tests)
+
+test-llama-gguf: prepare-test
+	TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="llama-gguf" --flake-attempts $(TEST_FLAKES) -v -r $(TEST_PATHS)
+
+test-tts: prepare-test
+	TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="tts" --flake-attempts $(TEST_FLAKES) -v -r $(TEST_PATHS)
+
+test-stablediffusion: prepare-test
+	TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="stablediffusion" --flake-attempts $(TEST_FLAKES) -v -r $(TEST_PATHS)
+
+test-stores: backend-assets/grpc/local-store
+	mkdir -p tests/integration/backend-assets/grpc
+	cp -f backend-assets/grpc/local-store tests/integration/backend-assets/grpc/
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="stores" --flake-attempts $(TEST_FLAKES) -v -r tests/integration
+
+test-container:
+	docker build --target requirements -t local-ai-test-container .
+	docker run -ti --rm --entrypoint /bin/bash -ti -v $(abspath ./):/build local-ai-test-container
 
 ## Help:
 help: ## Show this help.
@@ -228,3 +486,353 @@ help: ## Show this help.
 		if (/^[a-zA-Z_-]+:.*?##.*$$/) {printf "    ${YELLOW}%-20s${GREEN}%s${RESET}\n", $$1, $$2} \
 		else if (/^## .*$$/) {printf "  ${CYAN}%s${RESET}\n", substr($$1,4)} \
 		}' $(MAKEFILE_LIST)
+
+.PHONY: protogen
+protogen: protogen-go protogen-python
+
+.PHONY: protogen-clean
+protogen-clean: protogen-go-clean protogen-python-clean
+
+.PHONY: protogen-go
+protogen-go:
+	mkdir -p pkg/grpc/proto
+	protoc --experimental_allow_proto3_optional -Ibackend/ --go_out=pkg/grpc/proto/ --go_opt=paths=source_relative --go-grpc_out=pkg/grpc/proto/ --go-grpc_opt=paths=source_relative \
+    backend/backend.proto
+
+.PHONY: protogen-go-clean
+protogen-go-clean:
+	$(RM) pkg/grpc/proto/backend.pb.go pkg/grpc/proto/backend_grpc.pb.go
+	$(RM) bin/*
+
+.PHONY: protogen-python
+protogen-python: bark-protogen coqui-protogen diffusers-protogen exllama2-protogen rerankers-protogen transformers-protogen kokoro-protogen vllm-protogen faster-whisper-protogen
+
+.PHONY: protogen-python-clean
+protogen-python-clean: bark-protogen-clean coqui-protogen-clean diffusers-protogen-clean  exllama2-protogen-clean rerankers-protogen-clean transformers-protogen-clean kokoro-protogen-clean vllm-protogen-clean faster-whisper-protogen-clean
+
+.PHONY: bark-protogen
+bark-protogen:
+	$(MAKE) -C backend/python/bark protogen
+
+.PHONY: bark-protogen-clean
+bark-protogen-clean:
+	$(MAKE) -C backend/python/bark protogen-clean
+
+.PHONY: coqui-protogen
+coqui-protogen:
+	$(MAKE) -C backend/python/coqui protogen
+
+.PHONY: coqui-protogen-clean
+coqui-protogen-clean:
+	$(MAKE) -C backend/python/coqui protogen-clean
+
+.PHONY: diffusers-protogen
+diffusers-protogen:
+	$(MAKE) -C backend/python/diffusers protogen
+
+.PHONY: diffusers-protogen-clean
+diffusers-protogen-clean:
+	$(MAKE) -C backend/python/diffusers protogen-clean
+
+.PHONY: faster-whisper-protogen
+faster-whisper-protogen:
+	$(MAKE) -C backend/python/faster-whisper protogen
+
+.PHONY: faster-whisper-protogen-clean
+faster-whisper-protogen-clean:
+	$(MAKE) -C backend/python/faster-whisper protogen-clean
+
+.PHONY: exllama2-protogen
+exllama2-protogen:
+	$(MAKE) -C backend/python/exllama2 protogen
+
+.PHONY: exllama2-protogen-clean
+exllama2-protogen-clean:
+	$(MAKE) -C backend/python/exllama2 protogen-clean
+
+.PHONY: rerankers-protogen
+rerankers-protogen:
+	$(MAKE) -C backend/python/rerankers protogen
+
+.PHONY: rerankers-protogen-clean
+rerankers-protogen-clean:
+	$(MAKE) -C backend/python/rerankers protogen-clean
+
+.PHONY: transformers-protogen
+transformers-protogen:
+	$(MAKE) -C backend/python/transformers protogen
+
+.PHONY: transformers-protogen-clean
+transformers-protogen-clean:
+	$(MAKE) -C backend/python/transformers protogen-clean
+
+.PHONY: kokoro-protogen
+kokoro-protogen:
+	$(MAKE) -C backend/python/kokoro protogen
+
+.PHONY: kokoro-protogen-clean
+kokoro-protogen-clean:
+	$(MAKE) -C backend/python/kokoro protogen-clean
+
+.PHONY: vllm-protogen
+vllm-protogen:
+	$(MAKE) -C backend/python/vllm protogen
+
+.PHONY: vllm-protogen-clean
+vllm-protogen-clean:
+	$(MAKE) -C backend/python/vllm protogen-clean
+
+## GRPC
+# Note: it is duplicated in the Dockerfile
+prepare-extra-conda-environments: protogen-python
+	$(MAKE) -C backend/python/bark
+	$(MAKE) -C backend/python/coqui
+	$(MAKE) -C backend/python/diffusers
+	$(MAKE) -C backend/python/faster-whisper
+	$(MAKE) -C backend/python/vllm
+	$(MAKE) -C backend/python/rerankers
+	$(MAKE) -C backend/python/transformers
+	$(MAKE) -C backend/python/kokoro
+	$(MAKE) -C backend/python/exllama2
+
+prepare-test-extra: protogen-python
+	$(MAKE) -C backend/python/transformers
+	$(MAKE) -C backend/python/diffusers
+
+test-extra: prepare-test-extra
+	$(MAKE) -C backend/python/transformers test
+	$(MAKE) -C backend/python/diffusers test
+
+backend-assets:
+	mkdir -p backend-assets
+ifeq ($(BUILD_API_ONLY),true)
+	touch backend-assets/keep
+endif
+
+backend-assets/espeak-ng-data: sources/go-piper sources/go-piper/libpiper_binding.a
+	mkdir -p backend-assets/espeak-ng-data
+	@cp -rf sources/go-piper/piper-phonemize/pi/share/espeak-ng-data/. backend-assets/espeak-ng-data
+
+backend-assets/grpc: protogen-go replace
+	mkdir -p backend-assets/grpc
+
+backend-assets/grpc/huggingface: backend-assets/grpc
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/huggingface ./backend/go/llm/langchain/
+ifneq ($(UPX),)
+	$(UPX) backend-assets/grpc/huggingface
+endif
+
+backend/cpp/llama/llama.cpp:
+	LLAMA_VERSION=$(CPPLLAMA_VERSION) $(MAKE) -C backend/cpp/llama llama.cpp
+
+INSTALLED_PACKAGES=$(CURDIR)/backend/cpp/grpc/installed_packages
+INSTALLED_LIB_CMAKE=$(INSTALLED_PACKAGES)/lib/cmake
+ADDED_CMAKE_ARGS=-Dabsl_DIR=${INSTALLED_LIB_CMAKE}/absl \
+				 -DProtobuf_DIR=${INSTALLED_LIB_CMAKE}/protobuf \
+				 -Dutf8_range_DIR=${INSTALLED_LIB_CMAKE}/utf8_range \
+				 -DgRPC_DIR=${INSTALLED_LIB_CMAKE}/grpc \
+				 -DCMAKE_CXX_STANDARD_INCLUDE_DIRECTORIES=${INSTALLED_PACKAGES}/include
+build-llama-cpp-grpc-server:
+# Conditionally build grpc for the llama backend to use if needed
+ifdef BUILD_GRPC_FOR_BACKEND_LLAMA
+	$(MAKE) -C backend/cpp/grpc build
+	_PROTOBUF_PROTOC=${INSTALLED_PACKAGES}/bin/proto \
+	_GRPC_CPP_PLUGIN_EXECUTABLE=${INSTALLED_PACKAGES}/bin/grpc_cpp_plugin \
+	PATH="${INSTALLED_PACKAGES}/bin:${PATH}" \
+	CMAKE_ARGS="${CMAKE_ARGS} ${ADDED_CMAKE_ARGS}" \
+	LLAMA_VERSION=$(CPPLLAMA_VERSION) \
+	$(MAKE) -C backend/cpp/${VARIANT} grpc-server
+else
+	echo "BUILD_GRPC_FOR_BACKEND_LLAMA is not defined."
+	LLAMA_VERSION=$(CPPLLAMA_VERSION) $(MAKE) -C backend/cpp/${VARIANT} grpc-server
+endif
+
+# This target is for manually building a variant with-auto detected flags
+backend-assets/grpc/llama-cpp: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-cpp
+	$(MAKE) -C backend/cpp/llama-cpp purge
+	$(info ${GREEN}I llama-cpp build info:avx2${RESET})
+	$(MAKE) VARIANT="llama-cpp" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-cpp/grpc-server backend-assets/grpc/llama-cpp
+
+backend-assets/grpc/llama-cpp-avx2: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-avx2
+	$(MAKE) -C backend/cpp/llama-avx2 purge
+	$(info ${GREEN}I llama-cpp build info:avx2${RESET})
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=on -DGGML_AVX512=off -DGGML_FMA=on -DGGML_F16C=on" $(MAKE) VARIANT="llama-avx2" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-avx2/grpc-server backend-assets/grpc/llama-cpp-avx2
+
+backend-assets/grpc/llama-cpp-avx512: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-avx512
+	$(MAKE) -C backend/cpp/llama-avx512 purge
+	$(info ${GREEN}I llama-cpp build info:avx512${RESET})
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=on -DGGML_FMA=on -DGGML_F16C=on" $(MAKE) VARIANT="llama-avx512" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-avx512/grpc-server backend-assets/grpc/llama-cpp-avx512
+
+backend-assets/grpc/llama-cpp-avx: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-avx
+	$(MAKE) -C backend/cpp/llama-avx purge
+	$(info ${GREEN}I llama-cpp build info:avx${RESET})
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" $(MAKE) VARIANT="llama-avx" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-avx/grpc-server backend-assets/grpc/llama-cpp-avx
+
+backend-assets/grpc/llama-cpp-fallback: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-fallback
+	$(MAKE) -C backend/cpp/llama-fallback purge
+	$(info ${GREEN}I llama-cpp build info:fallback${RESET})
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=off -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" $(MAKE) VARIANT="llama-fallback" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-fallback/grpc-server backend-assets/grpc/llama-cpp-fallback
+
+backend-assets/grpc/llama-cpp-cuda: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-cuda
+	$(MAKE) -C backend/cpp/llama-cuda purge
+	$(info ${GREEN}I llama-cpp build info:cuda${RESET})
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off -DGGML_CUDA=ON" $(MAKE) VARIANT="llama-cuda" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-cuda/grpc-server backend-assets/grpc/llama-cpp-cuda
+
+backend-assets/grpc/llama-cpp-hipblas: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-hipblas
+	$(MAKE) -C backend/cpp/llama-hipblas purge
+	$(info ${GREEN}I llama-cpp build info:hipblas${RESET})
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=off -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" BUILD_TYPE="hipblas" $(MAKE) VARIANT="llama-hipblas" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-hipblas/grpc-server backend-assets/grpc/llama-cpp-hipblas
+
+backend-assets/grpc/llama-cpp-sycl_f16: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-sycl_f16
+	$(MAKE) -C backend/cpp/llama-sycl_f16 purge
+	$(info ${GREEN}I llama-cpp build info:sycl_f16${RESET})
+	BUILD_TYPE="sycl_f16" $(MAKE) VARIANT="llama-sycl_f16" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-sycl_f16/grpc-server backend-assets/grpc/llama-cpp-sycl_f16
+
+backend-assets/grpc/llama-cpp-sycl_f32: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-sycl_f32
+	$(MAKE) -C backend/cpp/llama-sycl_f32 purge
+	$(info ${GREEN}I llama-cpp build info:sycl_f32${RESET})
+	BUILD_TYPE="sycl_f32" $(MAKE) VARIANT="llama-sycl_f32" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-sycl_f32/grpc-server backend-assets/grpc/llama-cpp-sycl_f32
+
+backend-assets/grpc/llama-cpp-grpc: backend-assets/grpc backend/cpp/llama/llama.cpp
+	cp -rf backend/cpp/llama backend/cpp/llama-grpc
+	$(MAKE) -C backend/cpp/llama-grpc purge
+	$(info ${GREEN}I llama-cpp build info:grpc${RESET})
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_RPC=ON -DGGML_AVX=off -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" TARGET="--target grpc-server --target rpc-server" $(MAKE) VARIANT="llama-grpc" build-llama-cpp-grpc-server
+	cp -rfv backend/cpp/llama-grpc/grpc-server backend-assets/grpc/llama-cpp-grpc
+
+backend-assets/util/llama-cpp-rpc-server: backend-assets/grpc/llama-cpp-grpc
+	mkdir -p backend-assets/util/
+	cp -rf backend/cpp/llama-grpc/llama.cpp/build/bin/rpc-server backend-assets/util/llama-cpp-rpc-server
+
+backend-assets/grpc/bark-cpp: backend/go/bark/libbark.a backend-assets/grpc
+	CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/backend/go/bark/ LIBRARY_PATH=$(CURDIR)/backend/go/bark/ \
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/bark-cpp ./backend/go/bark/
+ifneq ($(UPX),)
+	$(UPX) backend-assets/grpc/bark-cpp
+endif
+
+backend-assets/grpc/piper: sources/go-piper sources/go-piper/libpiper_binding.a backend-assets/grpc backend-assets/espeak-ng-data
+	CGO_CXXFLAGS="$(PIPER_CGO_CXXFLAGS)" CGO_LDFLAGS="$(PIPER_CGO_LDFLAGS)" LIBRARY_PATH=$(CURDIR)/sources/go-piper \
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/piper ./backend/go/tts/
+ifneq ($(UPX),)
+	$(UPX) backend-assets/grpc/piper
+endif
+
+backend-assets/grpc/silero-vad: backend-assets/grpc backend-assets/lib/libonnxruntime.so.1
+	CGO_LDFLAGS="$(CGO_LDFLAGS)" CPATH="$(CPATH):$(CURDIR)/sources/onnxruntime/include/" LIBRARY_PATH=$(CURDIR)/backend-assets/lib \
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/silero-vad ./backend/go/vad/silero
+ifneq ($(UPX),)
+	$(UPX) backend-assets/grpc/silero-vad
+endif
+
+backend-assets/grpc/whisper: sources/whisper.cpp sources/whisper.cpp/libwhisper.a backend-assets/grpc
+	CGO_LDFLAGS="$(CGO_LDFLAGS) $(CGO_LDFLAGS_WHISPER)" C_INCLUDE_PATH="$(CURDIR)/sources/whisper.cpp/include:$(CURDIR)/sources/whisper.cpp/ggml/include" LIBRARY_PATH=$(CURDIR)/sources/whisper.cpp \
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/whisper ./backend/go/transcribe/whisper
+ifneq ($(UPX),)
+	$(UPX) backend-assets/grpc/whisper
+endif
+
+backend-assets/grpc/local-store: backend-assets/grpc
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/local-store ./backend/go/stores/
+ifneq ($(UPX),)
+	$(UPX) backend-assets/grpc/local-store
+endif
+
+grpcs: prepare $(GRPC_BACKENDS)
+
+DOCKER_IMAGE?=local-ai
+DOCKER_AIO_IMAGE?=local-ai-aio
+IMAGE_TYPE?=core
+BASE_IMAGE?=ubuntu:22.04
+
+docker:
+	docker build \
+		--build-arg BASE_IMAGE=$(BASE_IMAGE) \
+		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
+		--build-arg GO_TAGS="$(GO_TAGS)" \
+		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
+		--build-arg BUILD_TYPE=$(BUILD_TYPE) \
+		-t $(DOCKER_IMAGE) .
+
+docker-cuda11:
+	docker build \
+		--build-arg CUDA_MAJOR_VERSION=11 \
+		--build-arg CUDA_MINOR_VERSION=8 \
+		--build-arg BASE_IMAGE=$(BASE_IMAGE) \
+		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
+		--build-arg GO_TAGS="$(GO_TAGS)" \
+		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
+		--build-arg BUILD_TYPE=$(BUILD_TYPE) \
+		-t $(DOCKER_IMAGE)-cuda11 .
+
+docker-aio:
+	@echo "Building AIO image with base $(BASE_IMAGE) as $(DOCKER_AIO_IMAGE)"
+	docker build \
+		--build-arg BASE_IMAGE=$(BASE_IMAGE) \
+		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
+		-t $(DOCKER_AIO_IMAGE) -f Dockerfile.aio .
+
+docker-aio-all:
+	$(MAKE) docker-aio DOCKER_AIO_SIZE=cpu
+	$(MAKE) docker-aio DOCKER_AIO_SIZE=cpu
+
+docker-image-intel:
+	docker build \
+		--progress plain \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2025.1.0-0-devel-ubuntu24.04 \
+		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
+		--build-arg GO_TAGS="none" \
+		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
+		--build-arg BUILD_TYPE=sycl_f32 -t $(DOCKER_IMAGE) .
+
+docker-image-intel-xpu:
+	docker build \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2025.1.0-0-devel-ubuntu22.04 \
+		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
+		--build-arg GO_TAGS="none" \
+		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
+		--build-arg BUILD_TYPE=sycl_f32 -t $(DOCKER_IMAGE) .
+
+.PHONY: swagger
+swagger:
+	swag init -g core/http/app.go --output swagger
+
+.PHONY: gen-assets
+gen-assets:
+	$(GOCMD) run core/dependencies_manager/manager.go webui_static.yaml core/http/static/assets
+
+## Documentation
+docs/layouts/_default:
+	mkdir -p docs/layouts/_default
+
+docs/static/gallery.html: docs/layouts/_default
+	$(GOCMD) run ./.github/ci/modelslist.go ./gallery/index.yaml > docs/static/gallery.html
+
+docs/public: docs/layouts/_default docs/static/gallery.html
+	cd docs && hugo --minify
+
+docs-clean:
+	rm -rf docs/public
+	rm -rf docs/static/gallery.html
+
+.PHONY: docs
+docs: docs/static/gallery.html
+	cd docs && hugo serve

SECURITY.md

@@ -0,0 +1,42 @@
+# Security Policy
+
+## Introduction
+
+At LocalAI, we take the security of our software seriously. We understand the importance of protecting our community from vulnerabilities and are committed to ensuring the safety and security of our users.
+
+## Supported Versions
+
+We provide support and updates for certain versions of our software. The following table outlines which versions are currently supported with security updates:
+
+| Version | Supported          |
+| ------- | ------------------ |
+| > 2.0   | :white_check_mark: |
+| < 2.0   | :x:                |
+
+Please ensure that you are using a supported version to receive the latest security updates.
+
+## Reporting a Vulnerability
+
+We encourage the responsible disclosure of any security vulnerabilities. If you believe you've found a security issue in our software, we kindly ask you to follow the steps below to report it to us:
+
+1. **Email Us:** Send an email to [security@localai.io](mailto:security@localai.io) with a detailed report. Please do not disclose the vulnerability publicly or to any third parties before it has been addressed by us.
+
+2. **Expect a Response:** We aim to acknowledge receipt of vulnerability reports within 48 hours. Our security team will review your report and work closely with you to understand the impact and ensure a thorough investigation.
+
+3. **Collaboration:** If the vulnerability is accepted, we will work with you and our community to address the issue promptly. We'll keep you informed throughout the resolution process and may request additional information or collaboration.
+
+4. **Disclosure:** Once the vulnerability has been resolved, we encourage a coordinated disclosure. We believe in transparency and will work with you to ensure that our community is informed in a responsible manner.
+
+## Use of Third-Party Platforms
+
+As a Free and Open Source Software (FOSS) organization, we do not offer monetary bounties. However, researchers who wish to report vulnerabilities can also do so via [Huntr](https://huntr.dev/bounties), a platform that recognizes contributions to open source security.
+
+## Contact
+
+For any security-related inquiries beyond vulnerability reporting, please contact us at [security@localai.io](mailto:security@localai.io).
+
+## Acknowledgments
+
+We appreciate the efforts of those who contribute to the security of our project. Your responsible disclosure is invaluable to the safety and integrity of LocalAI.
+
+Thank you for helping us keep LocalAI secure.

aio/cpu/README.md

@@ -0,0 +1,5 @@
+## AIO CPU size
+
+Use this image with CPU-only.
+
+Please keep using only C++ backends so the base image is as small as possible (without CUDA, cuDNN, python, etc).

aio/cpu/embeddings.yaml

@@ -0,0 +1,12 @@
+embeddings: true
+name: text-embedding-ada-002
+parameters:
+  model: huggingface://bartowski/granite-embedding-107m-multilingual-GGUF/granite-embedding-107m-multilingual-f16.gguf
+
+usage: |
+    You can test this model with curl like this:
+
+    curl http://localhost:8080/embeddings -X POST -H "Content-Type: application/json" -d '{
+      "input": "Your text string goes here",
+      "model": "text-embedding-ada-002"
+    }'

aio/cpu/image-gen.yaml

@@ -0,0 +1,23 @@
+name: stablediffusion
+backend: stablediffusion-ggml
+cfg_scale: 4.5
+
+options:
+- sampler:euler
+parameters:
+  model: stable-diffusion-v1-5-pruned-emaonly-Q4_0.gguf
+step: 25
+
+download_files:
+- filename: "stable-diffusion-v1-5-pruned-emaonly-Q4_0.gguf"
+  sha256: "b8944e9fe0b69b36ae1b5bb0185b3a7b8ef14347fe0fa9af6c64c4829022261f"
+  uri: "huggingface://second-state/stable-diffusion-v1-5-GGUF/stable-diffusion-v1-5-pruned-emaonly-Q4_0.gguf"
+
+usage: |
+        curl http://localhost:8080/v1/images/generations \
+          -H "Content-Type: application/json" \
+          -d '{
+            "prompt": "<positive prompt>|<negative prompt>",
+            "step": 25,
+            "size": "512x512"
+          }'

aio/cpu/rerank.yaml

@@ -0,0 +1,27 @@
+name: jina-reranker-v1-base-en
+backend: rerankers
+parameters:
+  model: cross-encoder
+
+usage: |
+    You can test this model with curl like this:
+
+    curl http://localhost:8080/v1/rerank \
+      -H "Content-Type: application/json" \
+      -d '{
+      "model": "jina-reranker-v1-base-en",
+      "query": "Organic skincare products for sensitive skin",
+      "documents": [
+        "Eco-friendly kitchenware for modern homes",
+        "Biodegradable cleaning supplies for eco-conscious consumers",
+        "Organic cotton baby clothes for sensitive skin",
+        "Natural organic skincare range for sensitive skin",
+        "Tech gadgets for smart homes: 2024 edition",
+        "Sustainable gardening tools and compost solutions",
+        "Sensitive skin-friendly facial cleansers and toners",
+        "Organic food wraps and storage solutions",
+        "All-natural pet food for dogs with allergies",
+        "Yoga mats made from recycled materials"
+      ],
+      "top_n": 3
+    }'

aio/cpu/speech-to-text.yaml

@@ -0,0 +1,18 @@
+name: whisper-1
+backend: whisper
+parameters:
+  model: ggml-whisper-base.bin
+
+usage: |
+    ## example audio file
+    wget --quiet --show-progress -O gb1.ogg https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg
+
+    ## Send the example audio file to the transcriptions endpoint
+    curl http://localhost:8080/v1/audio/transcriptions \
+         -H "Content-Type: multipart/form-data" \
+         -F file="@$PWD/gb1.ogg" -F model="whisper-1"
+
+download_files:
+- filename: "ggml-whisper-base.bin"
+  sha256: "60ed5bc3dd14eea856493d334349b405782ddcaf0028d4b5df4088345fba2efe"
+  uri: "https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.bin"

aio/cpu/text-to-speech.yaml

@@ -0,0 +1,15 @@
+name: tts-1
+download_files:
+  - filename: voice-en-us-amy-low.tar.gz
+    uri: https://github.com/rhasspy/piper/releases/download/v0.0.2/voice-en-us-amy-low.tar.gz
+
+parameters:
+  model: en-us-amy-low.onnx
+
+usage: |
+    To test if this model works as expected, you can use the following curl command:
+
+    curl http://localhost:8080/tts -H "Content-Type: application/json" -d '{
+      "model":"voice-en-us-amy-low",
+      "input": "Hi, this is a test."
+    }'

aio/cpu/text-to-text.yaml

@@ -0,0 +1,57 @@
+context_size: 8192
+f16: true
+function:
+  grammar:
+    no_mixed_free_string: true
+    schema_type: llama3.1 # or JSON is supported too (json)
+  response_regex:
+  - <function=(?P<name>\w+)>(?P<arguments>.*)</function>
+mmap: true
+name: gpt-4
+parameters:
+  model: Hermes-3-Llama-3.2-3B-Q4_K_M.gguf
+stopwords:
+- <|im_end|>
+- <dummy32000>
+- <|eot_id|>
+- <|end_of_text|>
+template:
+  chat: |
+    <|begin_of_text|><|start_header_id|>system<|end_header_id|>
+    You are a helpful assistant<|eot_id|><|start_header_id|>user<|end_header_id|>
+    {{.Input }}
+    <|start_header_id|>assistant<|end_header_id|>
+  chat_message: |
+    <|start_header_id|>{{if eq .RoleName "assistant"}}assistant{{else if eq .RoleName "system"}}system{{else if eq .RoleName "tool"}}tool{{else if eq .RoleName "user"}}user{{end}}<|end_header_id|>
+    {{ if .FunctionCall -}}
+    {{ else if eq .RoleName "tool" -}}
+    The Function was executed and the response was:
+    {{ end -}}
+    {{ if .Content -}}
+    {{.Content -}}
+    {{ else if .FunctionCall -}}
+    {{ range .FunctionCall }}
+    [{{.FunctionCall.Name}}({{.FunctionCall.Arguments}})]
+    {{ end }}
+    {{ end -}}
+    <|eot_id|>
+  completion: |
+    {{.Input}}
+  function: |
+    <|start_header_id|>system<|end_header_id|>
+    You are an expert in composing functions. You are given a question and a set of possible functions.
+    Based on the question, you will need to make one or more function/tool calls to achieve the purpose.
+    If none of the functions can be used, point it out. If the given question lacks the parameters required by the function, also point it out. You should only return the function call in tools call sections.
+    If you decide to invoke any of the function(s), you MUST put it in the format as follows:
+    [func_name1(params_name1=params_value1,params_name2=params_value2,...),func_name2(params_name1=params_value1,params_name2=params_value2,...)]
+    You SHOULD NOT include any other text in the response.
+    Here is a list of functions in JSON format that you can invoke.
+    {{toJson .Functions}}
+    <|eot_id|><|start_header_id|>user<|end_header_id|>
+    {{.Input}}
+    <|eot_id|><|start_header_id|>assistant<|end_header_id|>
+
+download_files:
+- filename: Hermes-3-Llama-3.2-3B-Q4_K_M.gguf
+  sha256: 2e220a14ba4328fee38cf36c2c068261560f999fadb5725ce5c6d977cb5126b5
+  uri: huggingface://bartowski/Hermes-3-Llama-3.2-3B-GGUF/Hermes-3-Llama-3.2-3B-Q4_K_M.gguf

aio/cpu/vad.yaml

@@ -0,0 +1,8 @@
+backend: silero-vad
+name: silero-vad
+parameters:
+  model: silero-vad.onnx
+download_files:
+- filename: silero-vad.onnx
+  uri: https://huggingface.co/onnx-community/silero-vad/resolve/main/onnx/model.onnx
+  sha256: a4a068cd6cf1ea8355b84327595838ca748ec29a25bc91fc82e6c299ccdc5808

aio/cpu/vision.yaml

@@ -0,0 +1,49 @@
+context_size: 4096
+f16: true
+mmap: true
+mmproj: minicpm-v-2_6-mmproj-f16.gguf
+name: gpt-4o
+parameters:
+  model: minicpm-v-2_6-Q4_K_M.gguf
+stopwords:
+- <|im_end|>
+- <dummy32000>
+- </s>
+- <|endoftext|>
+template:
+  chat: |
+    {{.Input -}}
+    <|im_start|>assistant
+  chat_message: |
+    <|im_start|>{{ .RoleName }}
+    {{ if .FunctionCall -}}
+    Function call:
+    {{ else if eq .RoleName "tool" -}}
+    Function response:
+    {{ end -}}
+    {{ if .Content -}}
+    {{.Content }}
+    {{ end -}}
+    {{ if .FunctionCall -}}
+    {{toJson .FunctionCall}}
+    {{ end -}}<|im_end|>
+  completion: |
+    {{.Input}}
+  function: |
+    <|im_start|>system
+    You are a function calling AI model. You are provided with functions to execute. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools:
+    {{range .Functions}}
+    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
+    {{end}}
+    For each function call return a json object with function name and arguments
+    <|im_end|>
+    {{.Input -}}
+    <|im_start|>assistant
+
+download_files:
+- filename: minicpm-v-2_6-Q4_K_M.gguf
+  sha256: 3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1
+  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/ggml-model-Q4_K_M.gguf
+- filename: minicpm-v-2_6-mmproj-f16.gguf
+  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/mmproj-model-f16.gguf
+  sha256: 4485f68a0f1aa404c391e788ea88ea653c100d8e98fe572698f701e5809711fd

aio/entrypoint.sh

@@ -0,0 +1,138 @@
+#!/bin/bash
+
+echo "===> LocalAI All-in-One (AIO) container starting..."
+
+GPU_ACCELERATION=false
+GPU_VENDOR=""
+
+function check_intel() {
+    if lspci | grep -E 'VGA|3D' | grep -iq intel; then
+        echo "Intel GPU detected"
+        if [ -d /opt/intel ]; then
+            GPU_ACCELERATION=true
+            GPU_VENDOR=intel
+        else
+            echo "Intel GPU detected, but Intel GPU drivers are not installed. GPU acceleration will not be available."
+        fi
+    fi
+}
+
+function check_nvidia_wsl() {
+    if lspci | grep -E 'VGA|3D' | grep -iq "Microsoft Corporation Device 008e"; then
+        # We make the assumption this WSL2 cars is NVIDIA, then check for nvidia-smi
+        # Make sure the container was run with `--gpus all` as the only required parameter
+        echo "NVIDIA GPU detected via WSL2"
+        # nvidia-smi should be installed in the container
+        if nvidia-smi; then
+            GPU_ACCELERATION=true
+            GPU_VENDOR=nvidia
+        else
+            echo "NVIDIA GPU detected via WSL2, but nvidia-smi is not installed. GPU acceleration will not be available."
+        fi
+    fi
+}
+
+function check_amd() {
+    if lspci | grep -E 'VGA|3D' | grep -iq amd; then
+        echo "AMD GPU detected"
+        # Check if ROCm is installed
+        if [ -d /opt/rocm ]; then
+            GPU_ACCELERATION=true
+            GPU_VENDOR=amd
+        else
+            echo "AMD GPU detected, but ROCm is not installed. GPU acceleration will not be available."
+        fi
+    fi
+}
+
+function check_nvidia() {
+    if lspci | grep -E 'VGA|3D' | grep -iq nvidia; then
+        echo "NVIDIA GPU detected"
+        # nvidia-smi should be installed in the container
+        if nvidia-smi; then
+            GPU_ACCELERATION=true
+            GPU_VENDOR=nvidia
+        else
+            echo "NVIDIA GPU detected, but nvidia-smi is not installed. GPU acceleration will not be available."
+        fi
+    fi
+}
+
+function check_metal() {
+    if system_profiler SPDisplaysDataType | grep -iq 'Metal'; then
+        echo "Apple Metal supported GPU detected"
+        GPU_ACCELERATION=true
+        GPU_VENDOR=apple
+    fi
+}
+
+function detect_gpu() {
+    case "$(uname -s)" in
+        Linux)
+            check_nvidia
+            check_amd
+            check_intel
+            check_nvidia_wsl
+            ;;
+        Darwin)
+            check_metal
+            ;;
+    esac
+}
+
+function detect_gpu_size() {
+    # Attempting to find GPU memory size for NVIDIA GPUs
+    if [ "$GPU_ACCELERATION" = true ] && [ "$GPU_VENDOR" = "nvidia" ]; then
+        echo "NVIDIA GPU detected. Attempting to find memory size..."
+        # Using head -n 1 to get the total memory of the 1st NVIDIA GPU detected.
+        # If handling multiple GPUs is required in the future, this is the place to do it
+        nvidia_sm=$(nvidia-smi --query-gpu=memory.total --format=csv,noheader,nounits | head -n 1)
+        if [ ! -z "$nvidia_sm" ]; then
+            echo "Total GPU Memory: $nvidia_sm MiB"
+            # if bigger than 8GB, use 16GB
+            #if [ "$nvidia_sm" -gt 8192 ]; then
+            #    GPU_SIZE=gpu-16g
+            #else
+            GPU_SIZE=gpu-8g
+            #fi
+        else
+            echo "Unable to determine NVIDIA GPU memory size. Falling back to CPU."
+            GPU_SIZE=gpu-8g
+        fi
+    elif [ "$GPU_ACCELERATION" = true ] && [ "$GPU_VENDOR" = "intel" ]; then
+        GPU_SIZE=intel
+    # Default to a generic GPU size until we implement GPU size detection for non NVIDIA GPUs
+    elif [ "$GPU_ACCELERATION" = true ]; then
+        echo "Non-NVIDIA GPU detected. Specific GPU memory size detection is not implemented."
+        GPU_SIZE=gpu-8g
+
+    # default to cpu if GPU_SIZE is not set
+    else
+        echo "GPU acceleration is not enabled or supported. Defaulting to CPU."
+        GPU_SIZE=cpu
+    fi
+}
+
+function check_vars() {
+    if [ -z "$MODELS" ]; then
+        echo "MODELS environment variable is not set. Please set it to a comma-separated list of model YAML files to load."
+        exit 1
+    fi
+
+    if [ -z "$PROFILE" ]; then
+        echo "PROFILE environment variable is not set. Please set it to one of the following: cpu, gpu-8g, gpu-16g, apple"
+        exit 1
+    fi
+}
+
+detect_gpu
+detect_gpu_size
+
+PROFILE="${PROFILE:-$GPU_SIZE}" # default to cpu
+export MODELS="${MODELS:-/aio/${PROFILE}/embeddings.yaml,/aio/${PROFILE}/rerank.yaml,/aio/${PROFILE}/text-to-speech.yaml,/aio/${PROFILE}/image-gen.yaml,/aio/${PROFILE}/text-to-text.yaml,/aio/${PROFILE}/speech-to-text.yaml,/aio/${PROFILE}/vad.yaml,/aio/${PROFILE}/vision.yaml}"
+
+check_vars
+
+echo "===> Starting LocalAI[$PROFILE] with the following models: $MODELS"
+
+exec /build/entrypoint.sh "$@"

aio/gpu-8g/embeddings.yaml

@@ -0,0 +1,12 @@
+embeddings: true
+name: text-embedding-ada-002
+parameters:
+  model: huggingface://bartowski/granite-embedding-107m-multilingual-GGUF/granite-embedding-107m-multilingual-f16.gguf
+
+usage: |
+    You can test this model with curl like this:
+
+    curl http://localhost:8080/embeddings -X POST -H "Content-Type: application/json" -d '{
+      "input": "Your text string goes here",
+      "model": "text-embedding-ada-002"
+    }'

aio/gpu-8g/image-gen.yaml

@@ -0,0 +1,25 @@
+name: stablediffusion
+parameters:
+  model: DreamShaper_8_pruned.safetensors
+backend: diffusers
+step: 25
+f16: true
+
+diffusers:
+  pipeline_type: StableDiffusionPipeline
+  cuda: true
+  enable_parameters: "negative_prompt,num_inference_steps"
+  scheduler_type: "k_dpmpp_2m"
+
+download_files:
+- filename: DreamShaper_8_pruned.safetensors
+  uri: huggingface://Lykon/DreamShaper/DreamShaper_8_pruned.safetensors
+
+usage: |
+        curl http://localhost:8080/v1/images/generations \
+          -H "Content-Type: application/json" \
+          -d '{
+            "prompt": "<positive prompt>|<negative prompt>",
+            "step": 25,
+            "size": "512x512"
+          }'

aio/gpu-8g/rerank.yaml

@@ -0,0 +1,27 @@
+name: jina-reranker-v1-base-en
+backend: rerankers
+parameters:
+  model: cross-encoder
+
+usage: |
+    You can test this model with curl like this:
+
+    curl http://localhost:8080/v1/rerank \
+      -H "Content-Type: application/json" \
+      -d '{
+      "model": "jina-reranker-v1-base-en",
+      "query": "Organic skincare products for sensitive skin",
+      "documents": [
+        "Eco-friendly kitchenware for modern homes",
+        "Biodegradable cleaning supplies for eco-conscious consumers",
+        "Organic cotton baby clothes for sensitive skin",
+        "Natural organic skincare range for sensitive skin",
+        "Tech gadgets for smart homes: 2024 edition",
+        "Sustainable gardening tools and compost solutions",
+        "Sensitive skin-friendly facial cleansers and toners",
+        "Organic food wraps and storage solutions",
+        "All-natural pet food for dogs with allergies",
+        "Yoga mats made from recycled materials"
+      ],
+      "top_n": 3
+    }'

aio/gpu-8g/speech-to-text.yaml

@@ -0,0 +1,18 @@
+name: whisper-1
+backend: whisper
+parameters:
+  model: ggml-whisper-base.bin
+
+usage: |
+    ## example audio file
+    wget --quiet --show-progress -O gb1.ogg https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg
+
+    ## Send the example audio file to the transcriptions endpoint
+    curl http://localhost:8080/v1/audio/transcriptions \
+         -H "Content-Type: multipart/form-data" \
+         -F file="@$PWD/gb1.ogg" -F model="whisper-1"
+
+download_files:
+- filename: "ggml-whisper-base.bin"
+  sha256: "60ed5bc3dd14eea856493d334349b405782ddcaf0028d4b5df4088345fba2efe"
+  uri: "https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.bin"

aio/gpu-8g/text-to-speech.yaml

@@ -0,0 +1,15 @@
+name: tts-1
+download_files:
+  - filename: voice-en-us-amy-low.tar.gz
+    uri: https://github.com/rhasspy/piper/releases/download/v0.0.2/voice-en-us-amy-low.tar.gz
+
+parameters:
+  model: en-us-amy-low.onnx
+
+usage: |
+    To test if this model works as expected, you can use the following curl command:
+
+    curl http://localhost:8080/tts -H "Content-Type: application/json" -d '{
+      "model":"tts-1",
+      "input": "Hi, this is a test."
+    }'

aio/gpu-8g/text-to-text.yaml

@@ -0,0 +1,53 @@
+context_size: 4096
+f16: true
+function:
+  capture_llm_results:
+  - (?s)<Thought>(.*?)</Thought>
+  grammar:
+    properties_order: name,arguments
+  json_regex_match:
+  - (?s)<Output>(.*?)</Output>
+  replace_llm_results:
+  - key: (?s)<Thought>(.*?)</Thought>
+    value: ""
+mmap: true
+name: gpt-4
+parameters:
+  model: localai-functioncall-qwen2.5-7b-v0.5-q4_k_m.gguf
+stopwords:
+- <|im_end|>
+- <dummy32000>
+- </s>
+template:
+  chat: |
+    {{.Input -}}
+    <|im_start|>assistant
+  chat_message: |
+    <|im_start|>{{ .RoleName }}
+    {{ if .FunctionCall -}}
+    Function call:
+    {{ else if eq .RoleName "tool" -}}
+    Function response:
+    {{ end -}}
+    {{ if .Content -}}
+    {{.Content }}
+    {{ end -}}
+    {{ if .FunctionCall -}}
+    {{toJson .FunctionCall}}
+    {{ end -}}<|im_end|>
+  completion: |
+    {{.Input}}
+  function: |
+    <|im_start|>system
+    You are an AI assistant that executes function calls, and these are the tools at your disposal:
+    {{range .Functions}}
+    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
+    {{end}}
+    <|im_end|>
+    {{.Input -}}
+    <|im_start|>assistant
+
+download_files:
+- filename: localai-functioncall-phi-4-v0.3-q4_k_m.gguf
+  sha256: 23fee048ded2a6e2e1a7b6bbefa6cbf83068f194caa9552aecbaa00fec8a16d5
+  uri: huggingface://mudler/LocalAI-functioncall-phi-4-v0.3-Q4_K_M-GGUF/localai-functioncall-phi-4-v0.3-q4_k_m.gguf

aio/gpu-8g/vad.yaml

@@ -0,0 +1,8 @@
+backend: silero-vad
+name: silero-vad
+parameters:
+  model: silero-vad.onnx
+download_files:
+- filename: silero-vad.onnx
+  uri: https://huggingface.co/onnx-community/silero-vad/resolve/main/onnx/model.onnx
+  sha256: a4a068cd6cf1ea8355b84327595838ca748ec29a25bc91fc82e6c299ccdc5808

aio/gpu-8g/vision.yaml

@@ -0,0 +1,49 @@
+context_size: 4096
+f16: true
+mmap: true
+mmproj: minicpm-v-2_6-mmproj-f16.gguf
+name: gpt-4o
+parameters:
+  model: minicpm-v-2_6-Q4_K_M.gguf
+stopwords:
+- <|im_end|>
+- <dummy32000>
+- </s>
+- <|endoftext|>
+template:
+  chat: |
+    {{.Input -}}
+    <|im_start|>assistant
+  chat_message: |
+    <|im_start|>{{ .RoleName }}
+    {{ if .FunctionCall -}}
+    Function call:
+    {{ else if eq .RoleName "tool" -}}
+    Function response:
+    {{ end -}}
+    {{ if .Content -}}
+    {{.Content }}
+    {{ end -}}
+    {{ if .FunctionCall -}}
+    {{toJson .FunctionCall}}
+    {{ end -}}<|im_end|>
+  completion: |
+    {{.Input}}
+  function: |
+    <|im_start|>system
+    You are a function calling AI model. You are provided with functions to execute. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools:
+    {{range .Functions}}
+    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
+    {{end}}
+    For each function call return a json object with function name and arguments
+    <|im_end|>
+    {{.Input -}}
+    <|im_start|>assistant
+
+download_files:
+- filename: minicpm-v-2_6-Q4_K_M.gguf
+  sha256: 3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1
+  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/ggml-model-Q4_K_M.gguf
+- filename: minicpm-v-2_6-mmproj-f16.gguf
+  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/mmproj-model-f16.gguf
+  sha256: 4485f68a0f1aa404c391e788ea88ea653c100d8e98fe572698f701e5809711fd

aio/intel/embeddings.yaml

@@ -0,0 +1,12 @@
+embeddings: true
+name: text-embedding-ada-002
+parameters:
+  model: huggingface://bartowski/granite-embedding-107m-multilingual-GGUF/granite-embedding-107m-multilingual-f16.gguf
+
+usage: |
+    You can test this model with curl like this:
+
+    curl http://localhost:8080/embeddings -X POST -H "Content-Type: application/json" -d '{
+      "input": "Your text string goes here",
+      "model": "text-embedding-ada-002"
+    }'

aio/intel/image-gen.yaml

@@ -0,0 +1,20 @@
+name: stablediffusion
+parameters:
+  model: Lykon/dreamshaper-8
+backend: diffusers
+step: 25
+f16: true
+diffusers:
+  pipeline_type: StableDiffusionPipeline
+  cuda: true
+  enable_parameters: "negative_prompt,num_inference_steps"
+  scheduler_type: "k_dpmpp_2m"
+
+usage: |
+        curl http://localhost:8080/v1/images/generations \
+          -H "Content-Type: application/json" \
+          -d '{
+            "prompt": "<positive prompt>|<negative prompt>",
+            "step": 25,
+            "size": "512x512"
+          }'

aio/intel/rerank.yaml

@@ -0,0 +1,27 @@
+name: jina-reranker-v1-base-en
+backend: rerankers
+parameters:
+  model: cross-encoder
+
+usage: |
+    You can test this model with curl like this:
+
+    curl http://localhost:8080/v1/rerank \
+      -H "Content-Type: application/json" \
+      -d '{
+      "model": "jina-reranker-v1-base-en",
+      "query": "Organic skincare products for sensitive skin",
+      "documents": [
+        "Eco-friendly kitchenware for modern homes",
+        "Biodegradable cleaning supplies for eco-conscious consumers",
+        "Organic cotton baby clothes for sensitive skin",
+        "Natural organic skincare range for sensitive skin",
+        "Tech gadgets for smart homes: 2024 edition",
+        "Sustainable gardening tools and compost solutions",
+        "Sensitive skin-friendly facial cleansers and toners",
+        "Organic food wraps and storage solutions",
+        "All-natural pet food for dogs with allergies",
+        "Yoga mats made from recycled materials"
+      ],
+      "top_n": 3
+    }'

aio/intel/speech-to-text.yaml

@@ -0,0 +1,18 @@
+name: whisper-1
+backend: whisper
+parameters:
+  model: ggml-whisper-base.bin
+
+usage: |
+    ## example audio file
+    wget --quiet --show-progress -O gb1.ogg https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg
+
+    ## Send the example audio file to the transcriptions endpoint
+    curl http://localhost:8080/v1/audio/transcriptions \
+         -H "Content-Type: multipart/form-data" \
+         -F file="@$PWD/gb1.ogg" -F model="whisper-1"
+
+download_files:
+- filename: "ggml-whisper-base.bin"
+  sha256: "60ed5bc3dd14eea856493d334349b405782ddcaf0028d4b5df4088345fba2efe"
+  uri: "https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.bin"

aio/intel/text-to-speech.yaml

@@ -0,0 +1,15 @@
+name: tts-1
+download_files:
+  - filename: voice-en-us-amy-low.tar.gz
+    uri: https://github.com/rhasspy/piper/releases/download/v0.0.2/voice-en-us-amy-low.tar.gz
+
+parameters:
+  model: en-us-amy-low.onnx
+
+usage: |
+    To test if this model works as expected, you can use the following curl command:
+
+    curl http://localhost:8080/tts -H "Content-Type: application/json" -d '{
+      "model":"tts-1",
+      "input": "Hi, this is a test."
+    }'

aio/intel/text-to-text.yaml

@@ -0,0 +1,53 @@
+context_size: 4096
+f16: true
+function:
+  capture_llm_results:
+  - (?s)<Thought>(.*?)</Thought>
+  grammar:
+    properties_order: name,arguments
+  json_regex_match:
+  - (?s)<Output>(.*?)</Output>
+  replace_llm_results:
+  - key: (?s)<Thought>(.*?)</Thought>
+    value: ""
+mmap: true
+name: gpt-4
+parameters:
+  model: localai-functioncall-qwen2.5-7b-v0.5-q4_k_m.gguf
+stopwords:
+- <|im_end|>
+- <dummy32000>
+- </s>
+template:
+  chat: |
+    {{.Input -}}
+    <|im_start|>assistant
+  chat_message: |
+    <|im_start|>{{ .RoleName }}
+    {{ if .FunctionCall -}}
+    Function call:
+    {{ else if eq .RoleName "tool" -}}
+    Function response:
+    {{ end -}}
+    {{ if .Content -}}
+    {{.Content }}
+    {{ end -}}
+    {{ if .FunctionCall -}}
+    {{toJson .FunctionCall}}
+    {{ end -}}<|im_end|>
+  completion: |
+    {{.Input}}
+  function: |
+    <|im_start|>system
+    You are an AI assistant that executes function calls, and these are the tools at your disposal:
+    {{range .Functions}}
+    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
+    {{end}}
+    <|im_end|>
+    {{.Input -}}
+    <|im_start|>assistant
+
+download_files:
+- filename: localai-functioncall-phi-4-v0.3-q4_k_m.gguf
+  sha256: 23fee048ded2a6e2e1a7b6bbefa6cbf83068f194caa9552aecbaa00fec8a16d5
+  uri: huggingface://mudler/LocalAI-functioncall-phi-4-v0.3-Q4_K_M-GGUF/localai-functioncall-phi-4-v0.3-q4_k_m.gguf

aio/intel/vad.yaml

@@ -0,0 +1,8 @@
+backend: silero-vad
+name: silero-vad
+parameters:
+  model: silero-vad.onnx
+download_files:
+- filename: silero-vad.onnx
+  uri: https://huggingface.co/onnx-community/silero-vad/resolve/main/onnx/model.onnx
+  sha256: a4a068cd6cf1ea8355b84327595838ca748ec29a25bc91fc82e6c299ccdc5808

aio/intel/vision.yaml

@@ -0,0 +1,50 @@
+context_size: 4096
+f16: true
+mmap: true
+mmproj: minicpm-v-2_6-mmproj-f16.gguf
+name: gpt-4o
+parameters:
+  model: minicpm-v-2_6-Q4_K_M.gguf
+stopwords:
+- <|im_end|>
+- <dummy32000>
+- </s>
+- <|endoftext|>
+template:
+  chat: |
+    {{.Input -}}
+    <|im_start|>assistant
+  chat_message: |
+    <|im_start|>{{ .RoleName }}
+    {{ if .FunctionCall -}}
+    Function call:
+    {{ else if eq .RoleName "tool" -}}
+    Function response:
+    {{ end -}}
+    {{ if .Content -}}
+    {{.Content }}
+    {{ end -}}
+    {{ if .FunctionCall -}}
+    {{toJson .FunctionCall}}
+    {{ end -}}<|im_end|>
+  completion: |
+    {{.Input}}
+  function: |
+    <|im_start|>system
+    You are a function calling AI model. You are provided with functions to execute. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools:
+    {{range .Functions}}
+    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
+    {{end}}
+    For each function call return a json object with function name and arguments
+    <|im_end|>
+    {{.Input -}}
+    <|im_start|>assistant
+
+
+download_files:
+- filename: minicpm-v-2_6-Q4_K_M.gguf
+  sha256: 3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1
+  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/ggml-model-Q4_K_M.gguf
+- filename: minicpm-v-2_6-mmproj-f16.gguf
+  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/mmproj-model-f16.gguf
+  sha256: 4485f68a0f1aa404c391e788ea88ea653c100d8e98fe572698f701e5809711fd

api/api.go

@@ -1,100 +0,0 @@
-package api
-
-import (
-	"errors"
-
-	model "github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/gofiber/fiber/v2"
-	"github.com/gofiber/fiber/v2/middleware/cors"
-	"github.com/gofiber/fiber/v2/middleware/logger"
-	"github.com/gofiber/fiber/v2/middleware/recover"
-	"github.com/rs/zerolog"
-	"github.com/rs/zerolog/log"
-)
-
-func App(configFile string, loader *model.ModelLoader, uploadLimitMB, threads, ctxSize int, f16 bool, debug, disableMessage bool, imageDir string) *fiber.App {
-	zerolog.SetGlobalLevel(zerolog.InfoLevel)
-	if debug {
-		zerolog.SetGlobalLevel(zerolog.DebugLevel)
-	}
-
-	// Return errors as JSON responses
-	app := fiber.New(fiber.Config{
-		BodyLimit:             uploadLimitMB * 1024 * 1024, // this is the default limit of 4MB
-		DisableStartupMessage: disableMessage,
-		// Override default error handler
-		ErrorHandler: func(ctx *fiber.Ctx, err error) error {
-			// Status code defaults to 500
-			code := fiber.StatusInternalServerError
-
-			// Retrieve the custom status code if it's a *fiber.Error
-			var e *fiber.Error
-			if errors.As(err, &e) {
-				code = e.Code
-			}
-
-			// Send custom error page
-			return ctx.Status(code).JSON(
-				ErrorResponse{
-					Error: &APIError{Message: err.Error(), Code: code},
-				},
-			)
-		},
-	})
-
-	if debug {
-		app.Use(logger.New(logger.Config{
-			Format: "[${ip}]:${port} ${status} - ${method} ${path}\n",
-		}))
-	}
-
-	cm := make(ConfigMerger)
-	if err := cm.LoadConfigs(loader.ModelPath); err != nil {
-		log.Error().Msgf("error loading config files: %s", err.Error())
-	}
-
-	if configFile != "" {
-		if err := cm.LoadConfigFile(configFile); err != nil {
-			log.Error().Msgf("error loading config file: %s", err.Error())
-		}
-	}
-
-	if debug {
-		for k, v := range cm {
-			log.Debug().Msgf("Model: %s (config: %+v)", k, v)
-		}
-	}
-	// Default middleware config
-	app.Use(recover.New())
-	app.Use(cors.New())
-
-	// openAI compatible API endpoint
-	app.Post("/v1/chat/completions", chatEndpoint(cm, debug, loader, threads, ctxSize, f16))
-	app.Post("/chat/completions", chatEndpoint(cm, debug, loader, threads, ctxSize, f16))
-
-	app.Post("/v1/edits", editEndpoint(cm, debug, loader, threads, ctxSize, f16))
-	app.Post("/edits", editEndpoint(cm, debug, loader, threads, ctxSize, f16))
-
-	app.Post("/v1/completions", completionEndpoint(cm, debug, loader, threads, ctxSize, f16))
-	app.Post("/completions", completionEndpoint(cm, debug, loader, threads, ctxSize, f16))
-
-	app.Post("/v1/embeddings", embeddingsEndpoint(cm, debug, loader, threads, ctxSize, f16))
-	app.Post("/embeddings", embeddingsEndpoint(cm, debug, loader, threads, ctxSize, f16))
-
-	// /v1/engines/{engine_id}/embeddings
-
-	app.Post("/v1/engines/:model/embeddings", embeddingsEndpoint(cm, debug, loader, threads, ctxSize, f16))
-
-	app.Post("/v1/audio/transcriptions", transcriptEndpoint(cm, debug, loader, threads, ctxSize, f16))
-
-	app.Post("/v1/images/generations", imageEndpoint(cm, debug, loader, imageDir))
-
-	if imageDir != "" {
-		app.Static("/generated-images", imageDir)
-	}
-
-	app.Get("/v1/models", listModels(loader, cm))
-	app.Get("/models", listModels(loader, cm))
-
-	return app
-}

api/api_test.go

@@ -1,191 +0,0 @@
-package api_test
-
-import (
-	"context"
-	"os"
-	"path/filepath"
-	"runtime"
-
-	. "github.com/go-skynet/LocalAI/api"
-	"github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/gofiber/fiber/v2"
-	. "github.com/onsi/ginkgo/v2"
-	. "github.com/onsi/gomega"
-
-	openaigo "github.com/otiai10/openaigo"
-	"github.com/sashabaranov/go-openai"
-)
-
-var _ = Describe("API test", func() {
-
-	var app *fiber.App
-	var modelLoader *model.ModelLoader
-	var client *openai.Client
-	var client2 *openaigo.Client
-	Context("API query", func() {
-		BeforeEach(func() {
-			modelLoader = model.NewModelLoader(os.Getenv("MODELS_PATH"))
-			app = App("", modelLoader, 15, 1, 512, false, true, true, "")
-			go app.Listen("127.0.0.1:9090")
-
-			defaultConfig := openai.DefaultConfig("")
-			defaultConfig.BaseURL = "http://127.0.0.1:9090/v1"
-
-			client2 = openaigo.NewClient("")
-			client2.BaseURL = defaultConfig.BaseURL
-
-			// Wait for API to be ready
-			client = openai.NewClientWithConfig(defaultConfig)
-			Eventually(func() error {
-				_, err := client.ListModels(context.TODO())
-				return err
-			}, "2m").ShouldNot(HaveOccurred())
-		})
-		AfterEach(func() {
-			app.Shutdown()
-		})
-		It("returns the models list", func() {
-			models, err := client.ListModels(context.TODO())
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(models.Models)).To(Equal(10))
-		})
-		It("can generate completions", func() {
-			resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "testmodel", Prompt: "abcdedfghikl"})
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Choices)).To(Equal(1))
-			Expect(resp.Choices[0].Text).ToNot(BeEmpty())
-		})
-
-		It("can generate chat completions ", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "testmodel", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "abcdedfghikl"}}})
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Choices)).To(Equal(1))
-			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())
-		})
-
-		It("can generate completions from model configs", func() {
-			resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "gpt4all", Prompt: "abcdedfghikl"})
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Choices)).To(Equal(1))
-			Expect(resp.Choices[0].Text).ToNot(BeEmpty())
-		})
-
-		It("can generate chat completions from model configs", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "gpt4all-2", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "abcdedfghikl"}}})
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Choices)).To(Equal(1))
-			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())
-		})
-
-		It("returns errors", func() {
-			_, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "foomodel", Prompt: "abcdedfghikl"})
-			Expect(err).To(HaveOccurred())
-			Expect(err.Error()).To(ContainSubstring("error, status code: 500, message: could not load model - all backends returned error: 12 errors occurred:"))
-		})
-		It("transcribes audio", func() {
-			if runtime.GOOS != "linux" {
-				Skip("test supported only on linux")
-			}
-			resp, err := client.CreateTranscription(
-				context.Background(),
-				openai.AudioRequest{
-					Model:    openai.Whisper1,
-					FilePath: filepath.Join(os.Getenv("TEST_DIR"), "audio.wav"),
-				},
-			)
-			Expect(err).ToNot(HaveOccurred())
-			Expect(resp.Text).To(ContainSubstring("This is the Micro Machine Man presenting"))
-		})
-
-		It("calculate embeddings", func() {
-			if runtime.GOOS != "linux" {
-				Skip("test supported only on linux")
-			}
-			resp, err := client.CreateEmbeddings(
-				context.Background(),
-				openai.EmbeddingRequest{
-					Model: openai.AdaEmbeddingV2,
-					Input: []string{"sun", "cat"},
-				},
-			)
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Data[0].Embedding)).To(BeNumerically("==", 384))
-			Expect(len(resp.Data[1].Embedding)).To(BeNumerically("==", 384))
-
-			sunEmbedding := resp.Data[0].Embedding
-			resp2, err := client.CreateEmbeddings(
-				context.Background(),
-				openai.EmbeddingRequest{
-					Model: openai.AdaEmbeddingV2,
-					Input: []string{"sun"},
-				},
-			)
-			Expect(err).ToNot(HaveOccurred())
-			Expect(resp2.Data[0].Embedding).To(Equal(sunEmbedding))
-		})
-
-		Context("backends", func() {
-			It("runs rwkv", func() {
-				if runtime.GOOS != "linux" {
-					Skip("test supported only on linux")
-				}
-				resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "rwkv_test", Prompt: "Count up to five: one, two, three, four,"})
-				Expect(err).ToNot(HaveOccurred())
-				Expect(len(resp.Choices) > 0).To(BeTrue())
-				Expect(resp.Choices[0].Text).To(Equal(" five."))
-			})
-		})
-	})
-
-	Context("Config file", func() {
-		BeforeEach(func() {
-			modelLoader = model.NewModelLoader(os.Getenv("MODELS_PATH"))
-			app = App(os.Getenv("CONFIG_FILE"), modelLoader, 5, 1, 512, false, true, true, "")
-			go app.Listen("127.0.0.1:9090")
-
-			defaultConfig := openai.DefaultConfig("")
-			defaultConfig.BaseURL = "http://127.0.0.1:9090/v1"
-			client2 = openaigo.NewClient("")
-			client2.BaseURL = defaultConfig.BaseURL
-			// Wait for API to be ready
-			client = openai.NewClientWithConfig(defaultConfig)
-			Eventually(func() error {
-				_, err := client.ListModels(context.TODO())
-				return err
-			}, "2m").ShouldNot(HaveOccurred())
-		})
-		AfterEach(func() {
-			app.Shutdown()
-		})
-		It("can generate chat completions from config file", func() {
-
-			models, err := client.ListModels(context.TODO())
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(models.Models)).To(Equal(12))
-		})
-		It("can generate chat completions from config file", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "list1", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "abcdedfghikl"}}})
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Choices)).To(Equal(1))
-			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())
-		})
-		It("can generate chat completions from config file", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "list2", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "abcdedfghikl"}}})
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Choices)).To(Equal(1))
-			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())
-		})
-		It("can generate edit completions from config file", func() {
-			request := openaigo.EditCreateRequestBody{
-				Model:       "list2",
-				Instruction: "foo",
-				Input:       "bar",
-			}
-			resp, err := client2.CreateEdit(context.Background(), request)
-			Expect(err).ToNot(HaveOccurred())
-			Expect(len(resp.Choices)).To(Equal(1))
-			Expect(resp.Choices[0].Text).ToNot(BeEmpty())
-		})
-
-	})
-})

api/config.go

@@ -1,297 +0,0 @@
-package api
-
-import (
-	"encoding/json"
-	"fmt"
-	"io/ioutil"
-	"os"
-	"path/filepath"
-	"strings"
-
-	model "github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/gofiber/fiber/v2"
-	"github.com/rs/zerolog/log"
-	"gopkg.in/yaml.v3"
-)
-
-type Config struct {
-	OpenAIRequest               `yaml:"parameters"`
-	Name                        string            `yaml:"name"`
-	StopWords                   []string          `yaml:"stopwords"`
-	Cutstrings                  []string          `yaml:"cutstrings"`
-	TrimSpace                   []string          `yaml:"trimspace"`
-	ContextSize                 int               `yaml:"context_size"`
-	F16                         bool              `yaml:"f16"`
-	Threads                     int               `yaml:"threads"`
-	Debug                       bool              `yaml:"debug"`
-	Roles                       map[string]string `yaml:"roles"`
-	Embeddings                  bool              `yaml:"embeddings"`
-	Backend                     string            `yaml:"backend"`
-	TemplateConfig              TemplateConfig    `yaml:"template"`
-	MirostatETA                 float64           `yaml:"mirostat_eta"`
-	MirostatTAU                 float64           `yaml:"mirostat_tau"`
-	Mirostat                    int               `yaml:"mirostat"`
-	NGPULayers                  int               `yaml:"gpu_layers"`
-	ImageGenerationAssets       string            `yaml:"asset_dir"`
-	PromptStrings, InputStrings []string
-	InputToken                  [][]int
-}
-
-type TemplateConfig struct {
-	Completion string `yaml:"completion"`
-	Chat       string `yaml:"chat"`
-	Edit       string `yaml:"edit"`
-}
-
-type ConfigMerger map[string]Config
-
-func ReadConfigFile(file string) ([]*Config, error) {
-	c := &[]*Config{}
-	f, err := os.ReadFile(file)
-	if err != nil {
-		return nil, fmt.Errorf("cannot read config file: %w", err)
-	}
-	if err := yaml.Unmarshal(f, c); err != nil {
-		return nil, fmt.Errorf("cannot unmarshal config file: %w", err)
-	}
-
-	return *c, nil
-}
-
-func ReadConfig(file string) (*Config, error) {
-	c := &Config{}
-	f, err := os.ReadFile(file)
-	if err != nil {
-		return nil, fmt.Errorf("cannot read config file: %w", err)
-	}
-	if err := yaml.Unmarshal(f, c); err != nil {
-		return nil, fmt.Errorf("cannot unmarshal config file: %w", err)
-	}
-
-	return c, nil
-}
-
-func (cm ConfigMerger) LoadConfigFile(file string) error {
-	c, err := ReadConfigFile(file)
-	if err != nil {
-		return fmt.Errorf("cannot load config file: %w", err)
-	}
-
-	for _, cc := range c {
-		cm[cc.Name] = *cc
-	}
-	return nil
-}
-
-func (cm ConfigMerger) LoadConfig(file string) error {
-	c, err := ReadConfig(file)
-	if err != nil {
-		return fmt.Errorf("cannot read config file: %w", err)
-	}
-
-	cm[c.Name] = *c
-	return nil
-}
-
-func (cm ConfigMerger) LoadConfigs(path string) error {
-	files, err := ioutil.ReadDir(path)
-	if err != nil {
-		return err
-	}
-
-	for _, file := range files {
-		// Skip templates, YAML and .keep files
-		if !strings.Contains(file.Name(), ".yaml") {
-			continue
-		}
-		c, err := ReadConfig(filepath.Join(path, file.Name()))
-		if err == nil {
-			cm[c.Name] = *c
-		}
-	}
-
-	return nil
-}
-
-func updateConfig(config *Config, input *OpenAIRequest) {
-	if input.Echo {
-		config.Echo = input.Echo
-	}
-	if input.TopK != 0 {
-		config.TopK = input.TopK
-	}
-	if input.TopP != 0 {
-		config.TopP = input.TopP
-	}
-
-	if input.Temperature != 0 {
-		config.Temperature = input.Temperature
-	}
-
-	if input.Maxtokens != 0 {
-		config.Maxtokens = input.Maxtokens
-	}
-
-	switch stop := input.Stop.(type) {
-	case string:
-		if stop != "" {
-			config.StopWords = append(config.StopWords, stop)
-		}
-	case []interface{}:
-		for _, pp := range stop {
-			if s, ok := pp.(string); ok {
-				config.StopWords = append(config.StopWords, s)
-			}
-		}
-	}
-
-	if input.RepeatPenalty != 0 {
-		config.RepeatPenalty = input.RepeatPenalty
-	}
-
-	if input.Keep != 0 {
-		config.Keep = input.Keep
-	}
-
-	if input.Batch != 0 {
-		config.Batch = input.Batch
-	}
-
-	if input.F16 {
-		config.F16 = input.F16
-	}
-
-	if input.IgnoreEOS {
-		config.IgnoreEOS = input.IgnoreEOS
-	}
-
-	if input.Seed != 0 {
-		config.Seed = input.Seed
-	}
-
-	if input.Mirostat != 0 {
-		config.Mirostat = input.Mirostat
-	}
-
-	if input.MirostatETA != 0 {
-		config.MirostatETA = input.MirostatETA
-	}
-
-	if input.MirostatTAU != 0 {
-		config.MirostatTAU = input.MirostatTAU
-	}
-
-	switch inputs := input.Input.(type) {
-	case string:
-		if inputs != "" {
-			config.InputStrings = append(config.InputStrings, inputs)
-		}
-	case []interface{}:
-		for _, pp := range inputs {
-			switch i := pp.(type) {
-			case string:
-				config.InputStrings = append(config.InputStrings, i)
-			case []interface{}:
-				tokens := []int{}
-				for _, ii := range i {
-					tokens = append(tokens, int(ii.(float64)))
-				}
-				config.InputToken = append(config.InputToken, tokens)
-			}
-		}
-	}
-
-	switch p := input.Prompt.(type) {
-	case string:
-		config.PromptStrings = append(config.PromptStrings, p)
-	case []interface{}:
-		for _, pp := range p {
-			if s, ok := pp.(string); ok {
-				config.PromptStrings = append(config.PromptStrings, s)
-			}
-		}
-	}
-}
-func readInput(c *fiber.Ctx, loader *model.ModelLoader, randomModel bool) (string, *OpenAIRequest, error) {
-	input := new(OpenAIRequest)
-	// Get input data from the request body
-	if err := c.BodyParser(input); err != nil {
-		return "", nil, err
-	}
-
-	modelFile := input.Model
-
-	if c.Params("model") != "" {
-		modelFile = c.Params("model")
-	}
-
-	received, _ := json.Marshal(input)
-
-	log.Debug().Msgf("Request received: %s", string(received))
-
-	// Set model from bearer token, if available
-	bearer := strings.TrimLeft(c.Get("authorization"), "Bearer ")
-	bearerExists := bearer != "" && loader.ExistsInModelPath(bearer)
-
-	// If no model was specified, take the first available
-	if modelFile == "" && !bearerExists && randomModel {
-		models, _ := loader.ListModels()
-		if len(models) > 0 {
-			modelFile = models[0]
-			log.Debug().Msgf("No model specified, using: %s", modelFile)
-		} else {
-			log.Debug().Msgf("No model specified, returning error")
-			return "", nil, fmt.Errorf("no model specified")
-		}
-	}
-
-	// If a model is found in bearer token takes precedence
-	if bearerExists {
-		log.Debug().Msgf("Using model from bearer token: %s", bearer)
-		modelFile = bearer
-	}
-	return modelFile, input, nil
-}
-
-func readConfig(modelFile string, input *OpenAIRequest, cm ConfigMerger, loader *model.ModelLoader, debug bool, threads, ctx int, f16 bool) (*Config, *OpenAIRequest, error) {
-	// Load a config file if present after the model name
-	modelConfig := filepath.Join(loader.ModelPath, modelFile+".yaml")
-	if _, err := os.Stat(modelConfig); err == nil {
-		if err := cm.LoadConfig(modelConfig); err != nil {
-			return nil, nil, fmt.Errorf("failed loading model config (%s) %s", modelConfig, err.Error())
-		}
-	}
-
-	var config *Config
-	cfg, exists := cm[modelFile]
-	if !exists {
-		config = &Config{
-			OpenAIRequest: defaultRequest(modelFile),
-			ContextSize:   ctx,
-			Threads:       threads,
-			F16:           f16,
-			Debug:         debug,
-		}
-	} else {
-		config = &cfg
-	}
-
-	// Set the parameters for the language model prediction
-	updateConfig(config, input)
-
-	// Don't allow 0 as setting
-	if config.Threads == 0 {
-		if threads != 0 {
-			config.Threads = threads
-		} else {
-			config.Threads = 4
-		}
-	}
-
-	// Enforce debug flag if passed from CLI
-	if debug {
-		config.Debug = true
-	}
-
-	return config, input, nil
-}

api/openai.go

@@ -1,672 +0,0 @@
-package api
-
-import (
-	"bufio"
-	"bytes"
-	"encoding/base64"
-	"encoding/json"
-	"fmt"
-	"io"
-	"io/ioutil"
-	"net/http"
-	"os"
-	"path"
-	"path/filepath"
-	"strconv"
-	"strings"
-
-	"github.com/ggerganov/whisper.cpp/bindings/go/pkg/whisper"
-	model "github.com/go-skynet/LocalAI/pkg/model"
-	whisperutil "github.com/go-skynet/LocalAI/pkg/whisper"
-	llama "github.com/go-skynet/go-llama.cpp"
-	"github.com/gofiber/fiber/v2"
-	"github.com/rs/zerolog/log"
-	"github.com/valyala/fasthttp"
-)
-
-// APIError provides error information returned by the OpenAI API.
-type APIError struct {
-	Code    any     `json:"code,omitempty"`
-	Message string  `json:"message"`
-	Param   *string `json:"param,omitempty"`
-	Type    string  `json:"type"`
-}
-
-type ErrorResponse struct {
-	Error *APIError `json:"error,omitempty"`
-}
-
-type OpenAIUsage struct {
-	PromptTokens     int `json:"prompt_tokens"`
-	CompletionTokens int `json:"completion_tokens"`
-	TotalTokens      int `json:"total_tokens"`
-}
-
-type Item struct {
-	Embedding []float32 `json:"embedding"`
-	Index     int       `json:"index"`
-	Object    string    `json:"object,omitempty"`
-
-	// Images
-	URL     string `json:"url,omitempty"`
-	B64JSON string `json:"b64_json,omitempty"`
-}
-
-type OpenAIResponse struct {
-	Created int      `json:"created,omitempty"`
-	Object  string   `json:"object,omitempty"`
-	ID      string   `json:"id,omitempty"`
-	Model   string   `json:"model,omitempty"`
-	Choices []Choice `json:"choices,omitempty"`
-	Data    []Item   `json:"data,omitempty"`
-
-	Usage OpenAIUsage `json:"usage"`
-}
-
-type Choice struct {
-	Index        int      `json:"index,omitempty"`
-	FinishReason string   `json:"finish_reason,omitempty"`
-	Message      *Message `json:"message,omitempty"`
-	Delta        *Message `json:"delta,omitempty"`
-	Text         string   `json:"text,omitempty"`
-}
-
-type Message struct {
-	Role    string `json:"role,omitempty" yaml:"role"`
-	Content string `json:"content,omitempty" yaml:"content"`
-}
-
-type OpenAIModel struct {
-	ID     string `json:"id"`
-	Object string `json:"object"`
-}
-
-type OpenAIRequest struct {
-	Model string `json:"model" yaml:"model"`
-
-	// whisper
-	File     string `json:"file" validate:"required"`
-	Language string `json:"language"`
-	//whisper/image
-	ResponseFormat string `json:"response_format"`
-	// image
-	Size string `json:"size"`
-	// Prompt is read only by completion/image API calls
-	Prompt interface{} `json:"prompt" yaml:"prompt"`
-
-	// Edit endpoint
-	Instruction string      `json:"instruction" yaml:"instruction"`
-	Input       interface{} `json:"input" yaml:"input"`
-
-	Stop interface{} `json:"stop" yaml:"stop"`
-
-	// Messages is read only by chat/completion API calls
-	Messages []Message `json:"messages" yaml:"messages"`
-
-	Stream bool `json:"stream"`
-	Echo   bool `json:"echo"`
-	// Common options between all the API calls
-	TopP        float64 `json:"top_p" yaml:"top_p"`
-	TopK        int     `json:"top_k" yaml:"top_k"`
-	Temperature float64 `json:"temperature" yaml:"temperature"`
-	Maxtokens   int     `json:"max_tokens" yaml:"max_tokens"`
-
-	N int `json:"n"`
-
-	// Custom parameters - not present in the OpenAI API
-	Batch         int     `json:"batch" yaml:"batch"`
-	F16           bool    `json:"f16" yaml:"f16"`
-	IgnoreEOS     bool    `json:"ignore_eos" yaml:"ignore_eos"`
-	RepeatPenalty float64 `json:"repeat_penalty" yaml:"repeat_penalty"`
-	Keep          int     `json:"n_keep" yaml:"n_keep"`
-
-	MirostatETA float64 `json:"mirostat_eta" yaml:"mirostat_eta"`
-	MirostatTAU float64 `json:"mirostat_tau" yaml:"mirostat_tau"`
-	Mirostat    int     `json:"mirostat" yaml:"mirostat"`
-
-	Seed int `json:"seed" yaml:"seed"`
-
-	// Image (not supported by OpenAI)
-	Mode int `json:"mode"`
-	Step int `json:"step"`
-}
-
-func defaultRequest(modelFile string) OpenAIRequest {
-	return OpenAIRequest{
-		TopP:        0.7,
-		TopK:        80,
-		Maxtokens:   512,
-		Temperature: 0.9,
-		Model:       modelFile,
-	}
-}
-
-// https://platform.openai.com/docs/api-reference/completions
-func completionEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, threads, ctx int, f16 bool) func(c *fiber.Ctx) error {
-	return func(c *fiber.Ctx) error {
-
-		model, input, err := readInput(c, loader, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		config, input, err := readConfig(model, input, cm, loader, debug, threads, ctx, f16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		log.Debug().Msgf("Parameter Config: %+v", config)
-
-		templateFile := config.Model
-
-		if config.TemplateConfig.Completion != "" {
-			templateFile = config.TemplateConfig.Completion
-		}
-
-		var result []Choice
-		for _, i := range config.PromptStrings {
-			// A model can have a "file.bin.tmpl" file associated with a prompt template prefix
-			templatedInput, err := loader.TemplatePrefix(templateFile, struct {
-				Input string
-			}{Input: i})
-			if err == nil {
-				i = templatedInput
-				log.Debug().Msgf("Template found, input modified to: %s", i)
-			}
-
-			r, err := ComputeChoices(i, input, config, loader, func(s string, c *[]Choice) {
-				*c = append(*c, Choice{Text: s})
-			}, nil)
-			if err != nil {
-				return err
-			}
-
-			result = append(result, r...)
-		}
-
-		resp := &OpenAIResponse{
-			Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
-			Choices: result,
-			Object:  "text_completion",
-		}
-
-		jsonResult, _ := json.Marshal(resp)
-		log.Debug().Msgf("Response: %s", jsonResult)
-
-		// Return the prediction in the response body
-		return c.JSON(resp)
-	}
-}
-
-// https://platform.openai.com/docs/api-reference/embeddings
-func embeddingsEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, threads, ctx int, f16 bool) func(c *fiber.Ctx) error {
-	return func(c *fiber.Ctx) error {
-		model, input, err := readInput(c, loader, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		config, input, err := readConfig(model, input, cm, loader, debug, threads, ctx, f16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		log.Debug().Msgf("Parameter Config: %+v", config)
-		items := []Item{}
-
-		for i, s := range config.InputToken {
-			// get the model function to call for the result
-			embedFn, err := ModelEmbedding("", s, loader, *config)
-			if err != nil {
-				return err
-			}
-
-			embeddings, err := embedFn()
-			if err != nil {
-				return err
-			}
-			items = append(items, Item{Embedding: embeddings, Index: i, Object: "embedding"})
-		}
-
-		for i, s := range config.InputStrings {
-			// get the model function to call for the result
-			embedFn, err := ModelEmbedding(s, []int{}, loader, *config)
-			if err != nil {
-				return err
-			}
-
-			embeddings, err := embedFn()
-			if err != nil {
-				return err
-			}
-			items = append(items, Item{Embedding: embeddings, Index: i, Object: "embedding"})
-		}
-
-		resp := &OpenAIResponse{
-			Model:  input.Model, // we have to return what the user sent here, due to OpenAI spec.
-			Data:   items,
-			Object: "list",
-		}
-
-		jsonResult, _ := json.Marshal(resp)
-		log.Debug().Msgf("Response: %s", jsonResult)
-
-		// Return the prediction in the response body
-		return c.JSON(resp)
-	}
-}
-
-func chatEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, threads, ctx int, f16 bool) func(c *fiber.Ctx) error {
-
-	process := func(s string, req *OpenAIRequest, config *Config, loader *model.ModelLoader, responses chan OpenAIResponse) {
-		ComputeChoices(s, req, config, loader, func(s string, c *[]Choice) {}, func(s string) bool {
-			resp := OpenAIResponse{
-				Model:   req.Model, // we have to return what the user sent here, due to OpenAI spec.
-				Choices: []Choice{{Delta: &Message{Role: "assistant", Content: s}}},
-				Object:  "chat.completion.chunk",
-			}
-			log.Debug().Msgf("Sending goroutine: %s", s)
-
-			responses <- resp
-			return true
-		})
-		close(responses)
-	}
-	return func(c *fiber.Ctx) error {
-		model, input, err := readInput(c, loader, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		config, input, err := readConfig(model, input, cm, loader, debug, threads, ctx, f16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		log.Debug().Msgf("Parameter Config: %+v", config)
-
-		var predInput string
-
-		mess := []string{}
-		for _, i := range input.Messages {
-			var content string
-			r := config.Roles[i.Role]
-			if r != "" {
-				content = fmt.Sprint(r, " ", i.Content)
-			} else {
-				content = i.Content
-			}
-
-			mess = append(mess, content)
-		}
-
-		predInput = strings.Join(mess, "\n")
-
-		if input.Stream {
-			log.Debug().Msgf("Stream request received")
-			c.Context().SetContentType("text/event-stream")
-			//c.Response().Header.SetContentType(fiber.MIMETextHTMLCharsetUTF8)
-			//	c.Set("Content-Type", "text/event-stream")
-			c.Set("Cache-Control", "no-cache")
-			c.Set("Connection", "keep-alive")
-			c.Set("Transfer-Encoding", "chunked")
-		}
-
-		templateFile := config.Model
-
-		if config.TemplateConfig.Chat != "" {
-			templateFile = config.TemplateConfig.Chat
-		}
-
-		// A model can have a "file.bin.tmpl" file associated with a prompt template prefix
-		templatedInput, err := loader.TemplatePrefix(templateFile, struct {
-			Input string
-		}{Input: predInput})
-		if err == nil {
-			predInput = templatedInput
-			log.Debug().Msgf("Template found, input modified to: %s", predInput)
-		}
-
-		if input.Stream {
-			responses := make(chan OpenAIResponse)
-
-			go process(predInput, input, config, loader, responses)
-
-			c.Context().SetBodyStreamWriter(fasthttp.StreamWriter(func(w *bufio.Writer) {
-
-				for ev := range responses {
-					var buf bytes.Buffer
-					enc := json.NewEncoder(&buf)
-					enc.Encode(ev)
-
-					fmt.Fprintf(w, "event: data\n\n")
-					fmt.Fprintf(w, "data: %v\n\n", buf.String())
-					log.Debug().Msgf("Sending chunk: %s", buf.String())
-					w.Flush()
-				}
-
-				w.WriteString("event: data\n\n")
-				resp := &OpenAIResponse{
-					Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
-					Choices: []Choice{{FinishReason: "stop"}},
-				}
-				respData, _ := json.Marshal(resp)
-
-				w.WriteString(fmt.Sprintf("data: %s\n\n", respData))
-				w.Flush()
-			}))
-			return nil
-		}
-
-		result, err := ComputeChoices(predInput, input, config, loader, func(s string, c *[]Choice) {
-			*c = append(*c, Choice{Message: &Message{Role: "assistant", Content: s}})
-		}, nil)
-		if err != nil {
-			return err
-		}
-
-		resp := &OpenAIResponse{
-			Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
-			Choices: result,
-			Object:  "chat.completion",
-		}
-		respData, _ := json.Marshal(resp)
-		log.Debug().Msgf("Response: %s", respData)
-
-		// Return the prediction in the response body
-		return c.JSON(resp)
-	}
-}
-
-func editEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, threads, ctx int, f16 bool) func(c *fiber.Ctx) error {
-	return func(c *fiber.Ctx) error {
-		model, input, err := readInput(c, loader, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		config, input, err := readConfig(model, input, cm, loader, debug, threads, ctx, f16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		log.Debug().Msgf("Parameter Config: %+v", config)
-
-		templateFile := config.Model
-
-		if config.TemplateConfig.Edit != "" {
-			templateFile = config.TemplateConfig.Edit
-		}
-
-		var result []Choice
-		for _, i := range config.InputStrings {
-			// A model can have a "file.bin.tmpl" file associated with a prompt template prefix
-			templatedInput, err := loader.TemplatePrefix(templateFile, struct {
-				Input       string
-				Instruction string
-			}{Input: i})
-			if err == nil {
-				i = templatedInput
-				log.Debug().Msgf("Template found, input modified to: %s", i)
-			}
-
-			r, err := ComputeChoices(i, input, config, loader, func(s string, c *[]Choice) {
-				*c = append(*c, Choice{Text: s})
-			}, nil)
-			if err != nil {
-				return err
-			}
-
-			result = append(result, r...)
-		}
-
-		resp := &OpenAIResponse{
-			Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
-			Choices: result,
-			Object:  "edit",
-		}
-
-		jsonResult, _ := json.Marshal(resp)
-		log.Debug().Msgf("Response: %s", jsonResult)
-
-		// Return the prediction in the response body
-		return c.JSON(resp)
-	}
-}
-
-// https://platform.openai.com/docs/api-reference/images/create
-
-/*
-*
-
-	curl http://localhost:8080/v1/images/generations \
-	  -H "Content-Type: application/json" \
-	  -d '{
-	    "prompt": "A cute baby sea otter",
-	    "n": 1,
-	    "size": "512x512"
-	  }'
-
-*
-*/
-func imageEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, imageDir string) func(c *fiber.Ctx) error {
-	return func(c *fiber.Ctx) error {
-		m, input, err := readInput(c, loader, false)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		if m == "" {
-			m = model.StableDiffusionBackend
-		}
-		log.Debug().Msgf("Loading model: %+v", m)
-
-		config, input, err := readConfig(m, input, cm, loader, debug, 0, 0, false)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		log.Debug().Msgf("Parameter Config: %+v", config)
-
-		// XXX: Only stablediffusion is supported for now
-		if config.Backend == "" {
-			config.Backend = model.StableDiffusionBackend
-		}
-
-		sizeParts := strings.Split(input.Size, "x")
-		if len(sizeParts) != 2 {
-			return fmt.Errorf("Invalid value for 'size'")
-		}
-		width, err := strconv.Atoi(sizeParts[0])
-		if err != nil {
-			return fmt.Errorf("Invalid value for 'size'")
-		}
-		height, err := strconv.Atoi(sizeParts[1])
-		if err != nil {
-			return fmt.Errorf("Invalid value for 'size'")
-		}
-
-		b64JSON := false
-		if input.ResponseFormat == "b64_json" {
-			b64JSON = true
-		}
-
-		var result []Item
-		for _, i := range config.PromptStrings {
-			n := input.N
-			if input.N == 0 {
-				n = 1
-			}
-			for j := 0; j < n; j++ {
-				prompts := strings.Split(i, "|")
-				positive_prompt := prompts[0]
-				negative_prompt := ""
-				if len(prompts) > 1 {
-					negative_prompt = prompts[1]
-				}
-
-				mode := 0
-				step := 15
-
-				if input.Mode != 0 {
-					mode = input.Mode
-				}
-
-				if input.Step != 0 {
-					step = input.Step
-				}
-
-				tempDir := ""
-				if !b64JSON {
-					tempDir = imageDir
-				}
-				// Create a temporary file
-				outputFile, err := ioutil.TempFile(tempDir, "b64")
-				if err != nil {
-					return err
-				}
-				outputFile.Close()
-				output := outputFile.Name() + ".png"
-				// Rename the temporary file
-				err = os.Rename(outputFile.Name(), output)
-				if err != nil {
-					return err
-				}
-
-				baseURL := c.BaseURL()
-
-				fn, err := ImageGeneration(height, width, mode, step, input.Seed, positive_prompt, negative_prompt, output, loader, *config)
-				if err != nil {
-					return err
-				}
-				if err := fn(); err != nil {
-					return err
-				}
-
-				item := &Item{}
-
-				if b64JSON {
-					defer os.RemoveAll(output)
-					data, err := os.ReadFile(output)
-					if err != nil {
-						return err
-					}
-					item.B64JSON = base64.StdEncoding.EncodeToString(data)
-				} else {
-					base := filepath.Base(output)
-					item.URL = baseURL + "/generated-images/" + base
-				}
-
-				result = append(result, *item)
-			}
-		}
-
-		resp := &OpenAIResponse{
-			Data: result,
-		}
-
-		jsonResult, _ := json.Marshal(resp)
-		log.Debug().Msgf("Response: %s", jsonResult)
-
-		// Return the prediction in the response body
-		return c.JSON(resp)
-	}
-}
-
-// https://platform.openai.com/docs/api-reference/audio/create
-func transcriptEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, threads, ctx int, f16 bool) func(c *fiber.Ctx) error {
-	return func(c *fiber.Ctx) error {
-		m, input, err := readInput(c, loader, false)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		config, input, err := readConfig(m, input, cm, loader, debug, threads, ctx, f16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-		// retrieve the file data from the request
-		file, err := c.FormFile("file")
-		if err != nil {
-			return err
-		}
-		f, err := file.Open()
-		if err != nil {
-			return err
-		}
-		defer f.Close()
-
-		dir, err := os.MkdirTemp("", "whisper")
-
-		if err != nil {
-			return err
-		}
-		defer os.RemoveAll(dir)
-
-		dst := filepath.Join(dir, path.Base(file.Filename))
-		dstFile, err := os.Create(dst)
-		if err != nil {
-			return err
-		}
-
-		if _, err := io.Copy(dstFile, f); err != nil {
-			log.Debug().Msgf("Audio file copying error %+v - %+v - err %+v", file.Filename, dst, err)
-			return err
-		}
-
-		log.Debug().Msgf("Audio file copied to: %+v", dst)
-
-		whisperModel, err := loader.BackendLoader(model.WhisperBackend, config.Model, []llama.ModelOption{}, uint32(config.Threads))
-		if err != nil {
-			return err
-		}
-
-		if whisperModel == nil {
-			return fmt.Errorf("could not load whisper model")
-		}
-
-		w, ok := whisperModel.(whisper.Model)
-		if !ok {
-			return fmt.Errorf("loader returned non-whisper object")
-		}
-
-		tr, err := whisperutil.Transcript(w, dst, input.Language, uint(config.Threads))
-		if err != nil {
-			return err
-		}
-
-		log.Debug().Msgf("Trascribed: %+v", tr)
-		// TODO: handle different outputs here
-		return c.Status(http.StatusOK).JSON(fiber.Map{"text": tr})
-	}
-}
-
-func listModels(loader *model.ModelLoader, cm ConfigMerger) func(ctx *fiber.Ctx) error {
-	return func(c *fiber.Ctx) error {
-		models, err := loader.ListModels()
-		if err != nil {
-			return err
-		}
-		var mm map[string]interface{} = map[string]interface{}{}
-
-		dataModels := []OpenAIModel{}
-		for _, m := range models {
-			mm[m] = nil
-			dataModels = append(dataModels, OpenAIModel{ID: m, Object: "model"})
-		}
-
-		for k := range cm {
-			if _, exists := mm[k]; !exists {
-				dataModels = append(dataModels, OpenAIModel{ID: k, Object: "model"})
-			}
-		}
-
-		return c.JSON(struct {
-			Object string        `json:"object"`
-			Data   []OpenAIModel `json:"data"`
-		}{
-			Object: "list",
-			Data:   dataModels,
-		})
-	}
-}

api/prediction.go

@@ -1,561 +0,0 @@
-package api
-
-import (
-	"fmt"
-	"regexp"
-	"strings"
-	"sync"
-
-	"github.com/donomii/go-rwkv.cpp"
-	model "github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/go-skynet/LocalAI/pkg/stablediffusion"
-	"github.com/go-skynet/bloomz.cpp"
-	bert "github.com/go-skynet/go-bert.cpp"
-	gpt2 "github.com/go-skynet/go-gpt2.cpp"
-	llama "github.com/go-skynet/go-llama.cpp"
-	gpt4all "github.com/nomic-ai/gpt4all/gpt4all-bindings/golang"
-)
-
-// mutex still needed, see: https://github.com/ggerganov/llama.cpp/discussions/784
-var mutexMap sync.Mutex
-var mutexes map[string]*sync.Mutex = make(map[string]*sync.Mutex)
-
-func defaultLLamaOpts(c Config) []llama.ModelOption {
-	llamaOpts := []llama.ModelOption{}
-	if c.ContextSize != 0 {
-		llamaOpts = append(llamaOpts, llama.SetContext(c.ContextSize))
-	}
-	if c.F16 {
-		llamaOpts = append(llamaOpts, llama.EnableF16Memory)
-	}
-	if c.Embeddings {
-		llamaOpts = append(llamaOpts, llama.EnableEmbeddings)
-	}
-
-	if c.NGPULayers != 0 {
-		llamaOpts = append(llamaOpts, llama.SetGPULayers(c.NGPULayers))
-	}
-
-	return llamaOpts
-}
-
-func ImageGeneration(height, width, mode, step, seed int, positive_prompt, negative_prompt, dst string, loader *model.ModelLoader, c Config) (func() error, error) {
-	if c.Backend != model.StableDiffusionBackend {
-		return nil, fmt.Errorf("endpoint only working with stablediffusion models")
-	}
-	inferenceModel, err := loader.BackendLoader(c.Backend, c.ImageGenerationAssets, []llama.ModelOption{}, uint32(c.Threads))
-	if err != nil {
-		return nil, err
-	}
-
-	var fn func() error
-	switch model := inferenceModel.(type) {
-	case *stablediffusion.StableDiffusion:
-		fn = func() error {
-			return model.GenerateImage(height, width, mode, step, seed, positive_prompt, negative_prompt, dst)
-		}
-
-	default:
-		fn = func() error {
-			return fmt.Errorf("creation of images not supported by the backend")
-		}
-	}
-
-	return func() error {
-		// This is still needed, see: https://github.com/ggerganov/llama.cpp/discussions/784
-		mutexMap.Lock()
-		l, ok := mutexes[c.Backend]
-		if !ok {
-			m := &sync.Mutex{}
-			mutexes[c.Backend] = m
-			l = m
-		}
-		mutexMap.Unlock()
-		l.Lock()
-		defer l.Unlock()
-
-		return fn()
-	}, nil
-}
-
-func ModelEmbedding(s string, tokens []int, loader *model.ModelLoader, c Config) (func() ([]float32, error), error) {
-	if !c.Embeddings {
-		return nil, fmt.Errorf("endpoint disabled for this model by API configuration")
-	}
-
-	modelFile := c.Model
-
-	llamaOpts := defaultLLamaOpts(c)
-
-	var inferenceModel interface{}
-	var err error
-	if c.Backend == "" {
-		inferenceModel, err = loader.GreedyLoader(modelFile, llamaOpts, uint32(c.Threads))
-	} else {
-		inferenceModel, err = loader.BackendLoader(c.Backend, modelFile, llamaOpts, uint32(c.Threads))
-	}
-	if err != nil {
-		return nil, err
-	}
-
-	var fn func() ([]float32, error)
-	switch model := inferenceModel.(type) {
-	case *llama.LLama:
-		fn = func() ([]float32, error) {
-			predictOptions := buildLLamaPredictOptions(c)
-			if len(tokens) > 0 {
-				return model.TokenEmbeddings(tokens, predictOptions...)
-			}
-			return model.Embeddings(s, predictOptions...)
-		}
-	// bert embeddings
-	case *bert.Bert:
-		fn = func() ([]float32, error) {
-			if len(tokens) > 0 {
-				return model.TokenEmbeddings(tokens, bert.SetThreads(c.Threads))
-			}
-			return model.Embeddings(s, bert.SetThreads(c.Threads))
-		}
-	default:
-		fn = func() ([]float32, error) {
-			return nil, fmt.Errorf("embeddings not supported by the backend")
-		}
-	}
-
-	return func() ([]float32, error) {
-		// This is still needed, see: https://github.com/ggerganov/llama.cpp/discussions/784
-		mutexMap.Lock()
-		l, ok := mutexes[modelFile]
-		if !ok {
-			m := &sync.Mutex{}
-			mutexes[modelFile] = m
-			l = m
-		}
-		mutexMap.Unlock()
-		l.Lock()
-		defer l.Unlock()
-
-		embeds, err := fn()
-		if err != nil {
-			return embeds, err
-		}
-		// Remove trailing 0s
-		for i := len(embeds) - 1; i >= 0; i-- {
-			if embeds[i] == 0.0 {
-				embeds = embeds[:i]
-			} else {
-				break
-			}
-		}
-		return embeds, nil
-	}, nil
-}
-
-func buildLLamaPredictOptions(c Config) []llama.PredictOption {
-	// Generate the prediction using the language model
-	predictOptions := []llama.PredictOption{
-		llama.SetTemperature(c.Temperature),
-		llama.SetTopP(c.TopP),
-		llama.SetTopK(c.TopK),
-		llama.SetTokens(c.Maxtokens),
-		llama.SetThreads(c.Threads),
-	}
-
-	if c.Mirostat != 0 {
-		predictOptions = append(predictOptions, llama.SetMirostat(c.Mirostat))
-	}
-
-	if c.MirostatETA != 0 {
-		predictOptions = append(predictOptions, llama.SetMirostatETA(c.MirostatETA))
-	}
-
-	if c.MirostatTAU != 0 {
-		predictOptions = append(predictOptions, llama.SetMirostatTAU(c.MirostatTAU))
-	}
-
-	if c.Debug {
-		predictOptions = append(predictOptions, llama.Debug)
-	}
-
-	predictOptions = append(predictOptions, llama.SetStopWords(c.StopWords...))
-
-	if c.RepeatPenalty != 0 {
-		predictOptions = append(predictOptions, llama.SetPenalty(c.RepeatPenalty))
-	}
-
-	if c.Keep != 0 {
-		predictOptions = append(predictOptions, llama.SetNKeep(c.Keep))
-	}
-
-	if c.Batch != 0 {
-		predictOptions = append(predictOptions, llama.SetBatch(c.Batch))
-	}
-
-	if c.F16 {
-		predictOptions = append(predictOptions, llama.EnableF16KV)
-	}
-
-	if c.IgnoreEOS {
-		predictOptions = append(predictOptions, llama.IgnoreEOS)
-	}
-
-	if c.Seed != 0 {
-		predictOptions = append(predictOptions, llama.SetSeed(c.Seed))
-	}
-
-	return predictOptions
-}
-
-func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback func(string) bool) (func() (string, error), error) {
-	supportStreams := false
-	modelFile := c.Model
-
-	llamaOpts := defaultLLamaOpts(c)
-
-	var inferenceModel interface{}
-	var err error
-	if c.Backend == "" {
-		inferenceModel, err = loader.GreedyLoader(modelFile, llamaOpts, uint32(c.Threads))
-	} else {
-		inferenceModel, err = loader.BackendLoader(c.Backend, modelFile, llamaOpts, uint32(c.Threads))
-	}
-	if err != nil {
-		return nil, err
-	}
-
-	var fn func() (string, error)
-
-	switch model := inferenceModel.(type) {
-	case *rwkv.RwkvState:
-		supportStreams = true
-
-		fn = func() (string, error) {
-			stopWord := "\n"
-			if len(c.StopWords) > 0 {
-				stopWord = c.StopWords[0]
-			}
-
-			if err := model.ProcessInput(s); err != nil {
-				return "", err
-			}
-
-			response := model.GenerateResponse(c.Maxtokens, stopWord, float32(c.Temperature), float32(c.TopP), tokenCallback)
-
-			return response, nil
-		}
-	case *gpt2.GPTNeoX:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *gpt2.Replit:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *gpt2.Starcoder:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *gpt2.RedPajama:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *bloomz.Bloomz:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []bloomz.PredictOption{
-				bloomz.SetTemperature(c.Temperature),
-				bloomz.SetTopP(c.TopP),
-				bloomz.SetTopK(c.TopK),
-				bloomz.SetTokens(c.Maxtokens),
-				bloomz.SetThreads(c.Threads),
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, bloomz.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *gpt2.StableLM:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *gpt2.Dolly:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *gpt2.GPT2:
-		fn = func() (string, error) {
-			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
-			}
-
-			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
-			}
-
-			return model.Predict(
-				s,
-				predictOptions...,
-			)
-		}
-	case *gpt4all.Model:
-		supportStreams = true
-
-		fn = func() (string, error) {
-			if tokenCallback != nil {
-				model.SetTokenCallback(tokenCallback)
-			}
-
-			// Generate the prediction using the language model
-			predictOptions := []gpt4all.PredictOption{
-				gpt4all.SetTemperature(c.Temperature),
-				gpt4all.SetTopP(c.TopP),
-				gpt4all.SetTopK(c.TopK),
-				gpt4all.SetTokens(c.Maxtokens),
-			}
-
-			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt4all.SetBatch(c.Batch))
-			}
-
-			str, er := model.Predict(
-				s,
-				predictOptions...,
-			)
-			// Seems that if we don't free the callback explicitly we leave functions registered (that might try to send on closed channels)
-			// For instance otherwise the API returns: {"error":{"code":500,"message":"send on closed channel","type":""}}
-			// after a stream event has occurred
-			model.SetTokenCallback(nil)
-			return str, er
-		}
-	case *llama.LLama:
-		supportStreams = true
-		fn = func() (string, error) {
-
-			if tokenCallback != nil {
-				model.SetTokenCallback(tokenCallback)
-			}
-
-			predictOptions := buildLLamaPredictOptions(c)
-
-			str, er := model.Predict(
-				s,
-				predictOptions...,
-			)
-			// Seems that if we don't free the callback explicitly we leave functions registered (that might try to send on closed channels)
-			// For instance otherwise the API returns: {"error":{"code":500,"message":"send on closed channel","type":""}}
-			// after a stream event has occurred
-			model.SetTokenCallback(nil)
-			return str, er
-		}
-	}
-
-	return func() (string, error) {
-		// This is still needed, see: https://github.com/ggerganov/llama.cpp/discussions/784
-		mutexMap.Lock()
-		l, ok := mutexes[modelFile]
-		if !ok {
-			m := &sync.Mutex{}
-			mutexes[modelFile] = m
-			l = m
-		}
-		mutexMap.Unlock()
-		l.Lock()
-		defer l.Unlock()
-
-		res, err := fn()
-		if tokenCallback != nil && !supportStreams {
-			tokenCallback(res)
-		}
-		return res, err
-	}, nil
-}
-
-func ComputeChoices(predInput string, input *OpenAIRequest, config *Config, loader *model.ModelLoader, cb func(string, *[]Choice), tokenCallback func(string) bool) ([]Choice, error) {
-	result := []Choice{}
-
-	n := input.N
-
-	if input.N == 0 {
-		n = 1
-	}
-
-	// get the model function to call for the result
-	predFunc, err := ModelInference(predInput, loader, *config, tokenCallback)
-	if err != nil {
-		return result, err
-	}
-
-	for i := 0; i < n; i++ {
-		prediction, err := predFunc()
-		if err != nil {
-			return result, err
-		}
-
-		prediction = Finetune(*config, predInput, prediction)
-		cb(prediction, &result)
-
-		//result = append(result, Choice{Text: prediction})
-
-	}
-	return result, err
-}
-
-var cutstrings map[string]*regexp.Regexp = make(map[string]*regexp.Regexp)
-var mu sync.Mutex = sync.Mutex{}
-
-func Finetune(config Config, input, prediction string) string {
-	if config.Echo {
-		prediction = input + prediction
-	}
-
-	for _, c := range config.Cutstrings {
-		mu.Lock()
-		reg, ok := cutstrings[c]
-		if !ok {
-			cutstrings[c] = regexp.MustCompile(c)
-			reg = cutstrings[c]
-		}
-		mu.Unlock()
-		prediction = reg.ReplaceAllString(prediction, "")
-	}
-
-	for _, c := range config.TrimSpace {
-		prediction = strings.TrimSpace(strings.TrimPrefix(prediction, c))
-	}
-	return prediction
-
-}

assets.go

@@ -0,0 +1,6 @@
+package main
+
+import "embed"
+
+//go:embed backend-assets/*
+var backendAssets embed.FS

backend/backend.proto

@@ -0,0 +1,360 @@
+syntax = "proto3";
+
+option go_package = "github.com/go-skynet/LocalAI/pkg/grpc/proto";
+option java_multiple_files = true;
+option java_package = "io.skynet.localai.backend";
+option java_outer_classname = "LocalAIBackend";
+
+package backend;
+
+service Backend {
+  rpc Health(HealthMessage) returns (Reply) {}
+  rpc Predict(PredictOptions) returns (Reply) {}
+  rpc LoadModel(ModelOptions) returns (Result) {}
+  rpc PredictStream(PredictOptions) returns (stream Reply) {}
+  rpc Embedding(PredictOptions) returns (EmbeddingResult) {}
+  rpc GenerateImage(GenerateImageRequest) returns (Result) {}
+  rpc AudioTranscription(TranscriptRequest) returns (TranscriptResult) {}
+  rpc TTS(TTSRequest) returns (Result) {}
+  rpc SoundGeneration(SoundGenerationRequest) returns (Result) {}
+  rpc TokenizeString(PredictOptions) returns (TokenizationResponse) {}
+  rpc Status(HealthMessage) returns (StatusResponse) {}
+
+  rpc StoresSet(StoresSetOptions) returns (Result) {}
+  rpc StoresDelete(StoresDeleteOptions) returns (Result) {}
+  rpc StoresGet(StoresGetOptions) returns (StoresGetResult) {}
+  rpc StoresFind(StoresFindOptions) returns (StoresFindResult) {}
+
+  rpc Rerank(RerankRequest) returns (RerankResult) {}
+
+  rpc GetMetrics(MetricsRequest) returns (MetricsResponse);
+
+  rpc VAD(VADRequest) returns (VADResponse) {}
+}
+
+// Define the empty request
+message MetricsRequest {}
+
+message MetricsResponse {
+  int32 slot_id = 1;
+  string prompt_json_for_slot = 2;  // Stores the prompt as a JSON string.
+  float tokens_per_second = 3;
+  int32 tokens_generated = 4;
+  int32 prompt_tokens_processed = 5;
+}
+
+message RerankRequest {
+  string query = 1;
+  repeated string documents = 2;
+  int32 top_n = 3;
+}
+
+message RerankResult {
+  Usage usage = 1;
+  repeated DocumentResult results = 2;
+}
+
+message Usage {
+  int32 total_tokens = 1;
+  int32 prompt_tokens = 2;
+}
+
+message DocumentResult {
+  int32 index = 1;
+  string text = 2;
+  float relevance_score = 3;
+}
+
+message StoresKey {
+  repeated float Floats = 1;
+}
+
+message StoresValue {
+  bytes Bytes = 1;
+}
+
+message StoresSetOptions {
+  repeated StoresKey Keys = 1;
+  repeated StoresValue Values = 2;
+}
+
+message StoresDeleteOptions {
+  repeated StoresKey Keys = 1;
+}
+
+message StoresGetOptions {
+  repeated StoresKey Keys = 1;
+}
+
+message StoresGetResult {
+  repeated StoresKey Keys = 1;
+  repeated StoresValue Values = 2;
+}
+
+message StoresFindOptions {
+  StoresKey Key = 1;
+  int32 TopK = 2;
+}
+
+message StoresFindResult {
+  repeated StoresKey Keys = 1;
+  repeated StoresValue Values = 2;
+  repeated float Similarities = 3;
+}
+
+message HealthMessage {}
+
+// The request message containing the user's name.
+message PredictOptions {
+  string Prompt = 1;
+  int32 Seed = 2;
+  int32 Threads = 3;
+  int32 Tokens = 4;
+  int32 TopK = 5;
+  int32 Repeat = 6;
+  int32 Batch = 7;
+  int32 NKeep = 8;
+  float Temperature = 9;
+  float Penalty = 10;
+  bool F16KV = 11;
+  bool DebugMode = 12;
+  repeated string StopPrompts = 13;
+  bool IgnoreEOS = 14;
+  float TailFreeSamplingZ = 15;
+  float TypicalP = 16;
+  float FrequencyPenalty = 17;
+  float PresencePenalty = 18;
+  int32 Mirostat = 19;
+  float MirostatETA = 20;
+  float MirostatTAU = 21;
+  bool PenalizeNL = 22;
+  string LogitBias = 23;
+  bool MLock = 25;
+  bool MMap = 26;
+  bool PromptCacheAll = 27;
+  bool PromptCacheRO = 28;
+  string Grammar = 29;
+  string MainGPU = 30;
+  string TensorSplit = 31;
+  float TopP = 32;
+  string PromptCachePath = 33;
+  bool Debug = 34;
+  repeated int32 EmbeddingTokens = 35;
+  string Embeddings = 36;
+  float RopeFreqBase = 37;
+  float RopeFreqScale = 38;
+  float NegativePromptScale = 39;
+  string NegativePrompt = 40;
+  int32 NDraft = 41;
+  repeated string Images = 42;
+  bool UseTokenizerTemplate = 43;
+  repeated Message Messages = 44;
+  repeated string Videos = 45;
+  repeated string Audios = 46;
+  string CorrelationId = 47;
+}
+
+// The response message containing the result
+message Reply {
+  bytes message = 1;
+  int32 tokens = 2;
+  int32 prompt_tokens = 3;
+  double timing_prompt_processing = 4;
+  double timing_token_generation = 5;
+}
+
+message GrammarTrigger {
+  string word = 1;
+}
+
+message ModelOptions {
+  string Model = 1;
+  int32 ContextSize = 2;
+  int32 Seed = 3;
+  int32 NBatch = 4;
+  bool F16Memory = 5;
+  bool MLock = 6;
+  bool MMap = 7;
+  bool VocabOnly = 8;
+  bool LowVRAM = 9;
+  bool Embeddings = 10;
+  bool NUMA = 11;
+  int32 NGPULayers = 12;
+  string MainGPU = 13;
+  string TensorSplit = 14;
+  int32 Threads = 15;
+  string LibrarySearchPath = 16;
+  float RopeFreqBase = 17;
+  float RopeFreqScale = 18;
+  float RMSNormEps = 19;
+  int32 NGQA = 20;
+  string ModelFile = 21;
+
+
+
+  // Diffusers
+  string PipelineType = 26;
+  string SchedulerType = 27;
+  bool CUDA = 28;
+  float CFGScale = 29;
+  bool IMG2IMG = 30;
+  string CLIPModel = 31;
+  string CLIPSubfolder = 32;
+  int32 CLIPSkip = 33;
+  string ControlNet = 48;
+
+  string Tokenizer = 34;
+
+  // LLM (llama.cpp)
+  string LoraBase = 35;
+  string LoraAdapter = 36;
+  float LoraScale = 42;
+
+  bool NoMulMatQ = 37;
+  string DraftModel = 39;
+
+  string AudioPath = 38;
+
+  // vllm
+  string Quantization = 40;
+  float  GPUMemoryUtilization = 50;
+  bool   TrustRemoteCode = 51;
+  bool   EnforceEager = 52;
+  int32  SwapSpace = 53;
+  int32  MaxModelLen = 54;
+  int32  TensorParallelSize = 55;
+  string LoadFormat = 58;
+  bool   DisableLogStatus = 66;
+  string DType = 67;
+  int32  LimitImagePerPrompt = 68;
+  int32  LimitVideoPerPrompt = 69;
+  int32  LimitAudioPerPrompt = 70;
+
+  string MMProj = 41;
+
+  string RopeScaling = 43;
+  float YarnExtFactor = 44;
+  float YarnAttnFactor = 45;
+  float YarnBetaFast = 46;
+  float YarnBetaSlow = 47;
+
+  string Type = 49;
+
+  bool FlashAttention = 56;
+  bool NoKVOffload = 57;
+
+  string ModelPath = 59;
+
+  repeated string LoraAdapters = 60;
+  repeated float LoraScales = 61;
+
+  repeated string Options = 62;
+
+  string CacheTypeKey = 63;
+  string CacheTypeValue = 64;
+
+  repeated GrammarTrigger GrammarTriggers = 65;
+}
+
+message Result {
+  string message = 1;
+  bool success = 2;
+}
+
+message EmbeddingResult {
+  repeated float embeddings = 1;
+}
+
+message TranscriptRequest {
+  string dst = 2;
+  string language = 3;
+  uint32 threads = 4;
+  bool translate = 5;
+}
+
+message TranscriptResult {
+  repeated TranscriptSegment segments = 1;
+  string text = 2;
+}
+
+message TranscriptSegment {
+  int32 id = 1;
+  int64 start = 2;
+  int64 end = 3;
+  string text = 4;
+  repeated int32 tokens = 5;
+}
+
+message GenerateImageRequest {
+  int32 height = 1;
+  int32 width = 2;
+  int32 mode = 3;
+  int32 step = 4;
+  int32 seed = 5;
+  string positive_prompt = 6;
+  string negative_prompt = 7;
+  string dst = 8;
+  string src = 9;
+
+  // Diffusers
+  string EnableParameters = 10;
+  int32 CLIPSkip = 11;
+}
+
+message TTSRequest {
+  string text = 1;
+  string model = 2;
+  string dst = 3;
+  string voice = 4;
+  optional string language = 5;
+}
+
+message VADRequest {
+  repeated float audio = 1;
+}
+
+message VADSegment {
+  float start = 1;
+  float end = 2;
+}
+
+message VADResponse {
+  repeated VADSegment segments = 1;
+}
+
+message SoundGenerationRequest {
+  string text = 1;
+  string model = 2;
+  string dst = 3;
+  optional float duration = 4;
+  optional float temperature = 5;
+  optional bool sample = 6;
+  optional string src = 7;
+  optional int32 src_divisor = 8;
+}
+
+message TokenizationResponse {
+  int32 length = 1;
+  repeated int32 tokens = 2;
+}
+
+message MemoryUsageData {
+  uint64 total = 1;
+  map<string, uint64> breakdown = 2;
+}
+
+message StatusResponse {
+  enum State {
+    UNINITIALIZED = 0;
+    BUSY = 1;
+    READY = 2;
+    ERROR = -1;
+  }
+  State state = 1;
+  MemoryUsageData memory = 2;
+}
+
+message Message {
+  string role = 1;
+  string content = 2;
+}