tests: try to get logs

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

.github/check_and_update.py

@@ -1,80 +0,0 @@
-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()
-    if scan['hasUnsafeFile']:
-        return scan
-    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

@@ -14,14 +14,77 @@ function check_and_update_checksum() {
     idx="$5"
 
     # Download the file and calculate new checksum using Python
-    new_checksum=$(python3 ./.github/check_and_update.py $uri)
-    result=$?
+    new_checksum=$(python3 -c "
+import hashlib
+from huggingface_hub import hf_hub_download, get_paths_info
+import requests
+import sys
+import os
 
-    if [[ $result -eq 5 ]]; then
-        echo "Contaminated entry detected, deleting entry for $model_name..."
-        yq eval -i "del([$idx])" "$input_yaml"
-        return
-    fi
+uri = '$uri'
+file_name = uri.split('/')[-1]
+
+# Function to parse the URI and determine download method
+# 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()
+
+download_type, repo_id_or_url = parse_uri(uri)
+
+new_checksum =  None
+
+# Decide download method based on URI type
+if download_type == 'huggingface':
+    # 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)
+
+")
 
     if [[ "$new_checksum" == "" ]]; then
         echo "Error calculating checksum for $file_name. Skipping..."
@@ -31,7 +94,7 @@ function check_and_update_checksum() {
     echo "Checksum for $file_name: $new_checksum"
 
     # Compare and update the YAML file if checksums do not match
-    
+    result=$?
     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"

.github/dependabot.yml

@@ -1,10 +1,6 @@
 # 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:
@@ -27,111 +23,3 @@ updates:
     schedule:
       # Check for updates to GitHub Actions every weekday
       interval: "weekly"
-  - package-ecosystem: "pip"
-    directory: "/backend/python/autogptq"
-    schedule:
-      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/petals"
-    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/transformers-musicgen"
-    schedule:
-      interval: "weekly"
-  - package-ecosystem: "pip"
-    directory: "/backend/python/vall-e-x"
-    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/release.yml

@@ -13,9 +13,6 @@ changelog:
       labels:
         - bug
         - regression
-    - title: "🖧 P2P area"
-      labels:
-         - area/p2p
     - title: Exciting New Features 🎉
       labels:
         - Semver-Minor

.github/workflows/bump_deps.yaml

@@ -9,6 +9,9 @@ jobs:
       fail-fast: false
       matrix:
         include:
+          - repository: "go-skynet/go-llama.cpp"
+            variable: "GOLLAMA_VERSION"
+            branch: "master"
           - repository: "ggerganov/llama.cpp"
             variable: "CPPLLAMA_VERSION"
             branch: "master"
@@ -27,6 +30,9 @@ jobs:
           - repository: "go-skynet/bloomz.cpp"
             variable: "BLOOMZ_VERSION"
             branch: "main"
+          - repository: "nomic-ai/gpt4all"
+            variable: "GPT4ALL_VERSION"
+            branch: "main"
           - repository: "mudler/go-ggllm.cpp"
             variable: "GOGGLLM_VERSION"
             branch: "master"
@@ -48,7 +54,7 @@ jobs:
           token: ${{ secrets.UPDATE_BOT_TOKEN }}
           push-to-fork: ci-forks/LocalAI
           commit-message: ':arrow_up: Update ${{ matrix.repository }}'
-          title: 'chore: :arrow_up: Update ${{ matrix.repository }}'
+          title: ':arrow_up: Update ${{ matrix.repository }}'
           branch: "update/${{ matrix.variable }}"
           body: Bump of ${{ matrix.repository }} version
           signoff: true

.github/workflows/bump_docs.yaml

@@ -22,7 +22,7 @@ jobs:
           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 }}'
+          title: ':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

@@ -20,12 +20,12 @@ jobs:
         run: |
           sudo apt-get update
           sudo apt-get install -y pip wget
-          sudo pip install --upgrade pip
+          sudo pip install --upgrade pip 
           pip install huggingface_hub
       - name: 'Setup yq'
         uses: dcarbone/install-yq-action@v1.1.1
         with:
-          version: 'v4.44.2'
+          version: 'v4.43.1'
           download-compressed: true
           force: true
 

.github/workflows/dependabot_auto.yml

@@ -14,7 +14,7 @@ jobs:
     steps:
       - name: Dependabot metadata
         id: metadata
-        uses: dependabot/fetch-metadata@v2.2.0
+        uses: dependabot/fetch-metadata@v2.1.0
         with:
           github-token: "${{ secrets.GITHUB_TOKEN }}"
           skip-commit-verification: true

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

@@ -1,83 +0,0 @@
-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/generate_grpc_cache.yaml

@@ -75,7 +75,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: Cache GRPC
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v5
         with:
           builder: ${{ steps.buildx.outputs.name }}
           # The build-args MUST be an EXACT match between the image cache and other workflow steps that want to use that cache.
@@ -84,11 +84,11 @@ jobs:
           build-args: |
             GRPC_BASE_IMAGE=${{ matrix.grpc-base-image }}
             GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
-            GRPC_VERSION=v1.65.0
+            GRPC_VERSION=v1.64.0
           context: .
           file: ./Dockerfile
           cache-to: type=gha,ignore-error=true
           cache-from: type=gha
           target: grpc
           platforms: ${{ matrix.platforms }}
-          push: false
+          push: false

.github/workflows/generate_intel_image.yaml

@@ -15,7 +15,7 @@ jobs:
     strategy:
       matrix:
         include:
-          - base-image: intel/oneapi-basekit:2024.2.0-devel-ubuntu22.04
+          - base-image: intel/oneapi-basekit:2024.1.0-devel-ubuntu22.04
             runs-on: 'ubuntu-latest'
             platforms: 'linux/amd64'
     runs-on: ${{matrix.runs-on}}
@@ -46,7 +46,7 @@ jobs:
         uses: actions/checkout@v4
 
       - name: Cache Intel images
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v5
         with:
           builder: ${{ steps.buildx.outputs.name }}
           build-args: |

.github/workflows/image-pr.yml

@@ -35,19 +35,18 @@ jobs:
       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: ''
+            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: "4"
+            cuda-minor-version: "5"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-ffmpeg'
@@ -56,85 +55,85 @@ jobs:
             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: "4"
-          #   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"
+          - 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: "5"
+            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

@@ -39,7 +39,7 @@ jobs:
     strategy:
       # Pushing with all jobs in parallel
       # eats the bandwidth of all the nodes
-      max-parallel: ${{ github.event_name != 'pull_request' && 6 || 10 }}
+      max-parallel: ${{ github.event_name != 'pull_request' && 6 || 12 }}
       matrix:
         include:
           # Extra images
@@ -64,7 +64,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "11"
-            cuda-minor-version: "7"
+            cuda-minor-version: "8"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda11'
@@ -75,7 +75,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "4"
+            cuda-minor-version: "5"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12'
@@ -86,7 +86,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "11"
-            cuda-minor-version: "7"
+            cuda-minor-version: "8"
             platforms: 'linux/amd64'
             tag-latest: 'auto'
             tag-suffix: '-cublas-cuda11-ffmpeg'
@@ -100,7 +100,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "4"
+            cuda-minor-version: "5"
             platforms: 'linux/amd64'
             tag-latest: 'auto'
             tag-suffix: '-cublas-cuda12-ffmpeg'
@@ -257,7 +257,6 @@ jobs:
       quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
       quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
     strategy:
-      max-parallel: ${{ github.event_name != 'pull_request' && 2 || 4 }}
       matrix:
         include:
           - build-type: ''
@@ -274,7 +273,7 @@ jobs:
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "11"
-            cuda-minor-version: "7"
+            cuda-minor-version: "8"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda11-core'
@@ -285,7 +284,7 @@ jobs:
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "4"
+            cuda-minor-version: "5"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-core'
@@ -296,7 +295,7 @@ jobs:
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "11"
-            cuda-minor-version: "7"
+            cuda-minor-version: "8"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda11-ffmpeg-core'
@@ -307,7 +306,7 @@ jobs:
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "4"
+            cuda-minor-version: "5"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-ffmpeg-core'
@@ -317,12 +316,11 @@ jobs:
             base-image: "ubuntu:22.04"
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'vulkan'
-            platforms: 'linux/amd64'
+            platforms: 'linux/amd64,linux/arm64'
             tag-latest: 'false'
             tag-suffix: '-vulkan-ffmpeg-core'
-            latest-image: 'latest-vulkan-ffmpeg-core'
             ffmpeg: 'true'
             image-type: 'core'
             runs-on: 'arc-runner-set'
             base-image: "ubuntu:22.04"
-            makeflags: "--jobs=4 --output-sync=target"
+            makeflags: "--jobs=4 --output-sync=target"

.github/workflows/image_build.yml

@@ -23,7 +23,7 @@ on:
         type: string
       cuda-minor-version:
         description: 'CUDA minor version'
-        default: "4"
+        default: "5"
         type: string
       platforms:
         description: 'Platforms'
@@ -215,7 +215,7 @@ jobs:
           password: ${{ secrets.quayPassword }}
 
       - name: Build and push
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v5
         if: github.event_name != 'pull_request'
         with:
           builder: ${{ steps.buildx.outputs.name }}
@@ -232,7 +232,7 @@ jobs:
             BASE_IMAGE=${{ inputs.base-image }}
             GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
             GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
-            GRPC_VERSION=v1.65.0
+            GRPC_VERSION=v1.64.0
             MAKEFLAGS=${{ inputs.makeflags }}
           context: .
           file: ./Dockerfile
@@ -243,7 +243,7 @@ jobs:
           labels: ${{ steps.meta.outputs.labels }}
 ### Start testing image
       - name: Build and push
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v5
         if: github.event_name == 'pull_request'
         with:
           builder: ${{ steps.buildx.outputs.name }}
@@ -260,7 +260,7 @@ jobs:
             BASE_IMAGE=${{ inputs.base-image }}
             GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
             GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
-            GRPC_VERSION=v1.65.0
+            GRPC_VERSION=v1.64.0
             MAKEFLAGS=${{ inputs.makeflags }}
           context: .
           file: ./Dockerfile
@@ -276,7 +276,7 @@ jobs:
 ## End testing image
       - name: Build and push AIO image
         if: inputs.aio != ''
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v5
         with:
           builder: ${{ steps.buildx.outputs.name }}
           build-args: |
@@ -291,7 +291,7 @@ jobs:
 
       - name: Build and push AIO image (dockerhub)
         if: inputs.aio != ''
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v5
         with:
           builder: ${{ steps.buildx.outputs.name }}
           build-args: |

.github/workflows/notify-models.yaml

@@ -1,168 +0,0 @@
-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: hermes-2-theta-llama-3-8b
-    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: '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"
-    - 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@v3.18
-      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: hermes-2-theta-llama-3-8b
-    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@v2.7.0
-      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@v3.18
-      with:
-        detached: true
-        connect-timeout-seconds: 180
-        limit-access-to-actor: true

.github/workflows/notify-releases.yaml

@@ -1,63 +0,0 @@
-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: ${{ github.event.release.tag_name }}
-    steps:
-    - uses: mudler/localai-github-action@v1
-      with:
-        model: 'hermes-2-theta-llama-3-8b' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
-    - name: Summarize
-      id: summarize
-      run: |
-            input="$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

@@ -1,28 +0,0 @@
-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

@@ -1,15 +1,11 @@
 name: Build and Release
 
 on:
-  push:
-    branches:
-      - master
-    tags:
-      - 'v*'
-  pull_request:
+- push
+- pull_request
 
 env:
-  GRPC_VERSION: v1.65.0
+  GRPC_VERSION: v1.64.0
 
 permissions:
   contents: write
@@ -35,8 +31,8 @@ jobs:
       - name: Dependencies
         run: |
           sudo apt-get update
-          sudo apt-get install build-essential ffmpeg protobuf-compiler ccache gawk
-          sudo apt-get install -qy binutils-aarch64-linux-gnu gcc-aarch64-linux-gnu g++-aarch64-linux-gnu libgmock-dev
+          sudo apt-get install build-essential ffmpeg protobuf-compiler ccache
+          sudo apt-get install -qy binutils-aarch64-linux-gnu gcc-aarch64-linux-gnu g++-aarch64-linux-gnu
       - 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
@@ -44,7 +40,7 @@ jobs:
           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
+          CUDA_VERSION: 12-5
       - name: Cache grpc
         id: cache-grpc
         uses: actions/cache@v4
@@ -56,8 +52,7 @@ jobs:
         run: |
 
           git clone --recurse-submodules -b ${{ env.GRPC_VERSION }} --depth 1 --shallow-submodules https://github.com/grpc/grpc && \
-          cd grpc && sed -i "216i\  TESTONLY" "third_party/abseil-cpp/absl/container/CMakeLists.txt" && mkdir -p cmake/build && \
-          cd cmake/build && cmake -DgRPC_INSTALL=ON \
+          cd grpc && mkdir -p cmake/build && cd cmake/build && cmake -DgRPC_INSTALL=ON \
             -DgRPC_BUILD_TESTS=OFF \
             ../.. && sudo make --jobs 5 --output-sync=target
       - name: Install gRPC
@@ -101,13 +96,14 @@ jobs:
           CROSS_TOOLCHAIN=/usr/$GNU_HOST
           CROSS_STAGING_PREFIX=$CROSS_TOOLCHAIN/stage
           CMAKE_CROSS_TOOLCHAIN=/tmp/arm.toolchain.cmake
-          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
-          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
           export PATH=$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
+          GO_TAGS=p2p \
           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 \
@@ -151,7 +147,7 @@ jobs:
       - name: Dependencies
         run: |
           sudo apt-get update
-          sudo apt-get install -y wget curl build-essential ffmpeg protobuf-compiler ccache gawk cmake libgmock-dev
+          sudo apt-get install -y wget curl build-essential ffmpeg protobuf-compiler ccache cmake
       - 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
@@ -165,7 +161,7 @@ jobs:
           sudo apt-get update
           sudo apt-get install -y cuda-nvcc-${CUDA_VERSION} libcublas-dev-${CUDA_VERSION}
         env:
-          CUDA_VERSION: 12-5
+          CUDA_VERSION: 12-3
       - name: "Install Hipblas"
         env:
           ROCM_VERSION: "6.1"
@@ -201,8 +197,7 @@ jobs:
         if: steps.cache-grpc.outputs.cache-hit != 'true'
         run: |
           git clone --recurse-submodules -b ${{ env.GRPC_VERSION }} --depth 1 --shallow-submodules https://github.com/grpc/grpc && \
-          cd grpc && sed -i "216i\  TESTONLY" "third_party/abseil-cpp/absl/container/CMakeLists.txt" && mkdir -p cmake/build && \
-          cd cmake/build && cmake -DgRPC_INSTALL=ON \
+          cd grpc && mkdir -p cmake/build && cd cmake/build && cmake -DgRPC_INSTALL=ON \
             -DgRPC_BUILD_TESTS=OFF \
             ../.. && sudo make --jobs 5 --output-sync=target
       - name: Install gRPC
@@ -212,14 +207,15 @@ jobs:
       - name: Build
         id: build
         run: |
-          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
-          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
           export PATH=$PATH:$GOPATH/bin
           export PATH=/usr/local/cuda/bin:$PATH
           export PATH=/opt/rocm/bin:$PATH
           source /opt/intel/oneapi/setvars.sh
           sudo cp /lib64/ld-linux-x86-64.so.2 ld.so
-          BACKEND_LIBS="./ld.so ./sources/go-piper/piper/build/fi/lib/libfmt.a ./sources/go-piper/piper-phonemize/pi/lib/libonnxruntime.so.1.14.1 ./sources/go-piper/piper-phonemize/pi/src/libespeak-ng/libespeak-ng.so /usr/lib/x86_64-linux-gnu/libdl.so.2 /usr/lib/x86_64-linux-gnu/librt.so.1 /usr/lib/x86_64-linux-gnu/libpthread.so.0 ./sources/go-piper/piper-phonemize/pi/lib/libpiper_phonemize.so.1 ./sources/go-piper/piper/build/si/lib/libspdlog.a ./sources/go-piper/espeak/ei/lib/libucd.so" \
+          GO_TAGS=p2p \
+          BACKEND_LIBS="./ld.so /usr/lib/x86_64-linux-gnu/libstdc++.so.6 /usr/lib/x86_64-linux-gnu/libm.so.6 /usr/lib/x86_64-linux-gnu/libgcc_s.so.1 /usr/lib/x86_64-linux-gnu/libc.so.6 /usr/lib/x86_64-linux-gnu/libgomp.so.1" \
           make -j4 dist
       - uses: actions/upload-artifact@v4
         with:
@@ -253,8 +249,8 @@ jobs:
         run: |
           sudo apt-get update
           sudo apt-get install -y --no-install-recommends libopencv-dev protobuf-compiler ccache
-          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
-          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
       - name: Build stablediffusion
         run: |
           export PATH=$PATH:$GOPATH/bin
@@ -273,8 +269,8 @@ jobs:
           files: |
             release/*
 
-  build-macOS-x86_64:
-    runs-on: macos-13
+  build-macOS-arm64:
+    runs-on: macos-14
     steps:
       - name: Clone
         uses: actions/checkout@v4
@@ -296,49 +292,7 @@ jobs:
           export CPLUS_INCLUDE_PATH=/usr/local/include
           export PATH=$PATH:$GOPATH/bin
 
-          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: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.18
-        with:
-          detached: true
-          connect-timeout-seconds: 180
-          limit-access-to-actor: true
-
-  build-macOS-arm64:
-    runs-on: macos-14
-    steps:
-      - name: Clone
-        uses: actions/checkout@v4
-        with:
-          submodules: true
-      - uses: actions/setup-go@v5
-        with:
-          go-version: '1.21.x'
-          cache: false
-      - name: Dependencies
-        run: |
-          brew install protobuf grpc
-          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
-          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
-      - name: Build
-        id: build
-        run: |
-          export C_INCLUDE_PATH=/usr/local/include
-          export CPLUS_INCLUDE_PATH=/usr/local/include
-          export PATH=$PATH:$GOPATH/bin
-
-          make dist
+          BACKEND_LIBS="$(ls /opt/homebrew/opt/grpc/lib/*.dylib /opt/homebrew/opt/re2/lib/*.dylib /opt/homebrew/opt/openssl@3/lib/*.dylib /opt/homebrew/opt/protobuf/lib/*.dylib /opt/homebrew/opt/abseil/lib/*.dylib | xargs)" GO_TAGS=p2p make dist
       - uses: actions/upload-artifact@v4
         with:
           name: LocalAI-MacOS-arm64

.github/workflows/test-extra.yml

@@ -19,7 +19,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -29,8 +29,8 @@ jobs:
           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
-
+          pip install --user grpcio-tools==1.64.0
+          
       - name: Test transformers
         run: |
            make --jobs=5 --output-sync=target -C backend/python/transformers
@@ -41,7 +41,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -51,8 +51,8 @@ jobs:
           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
-
+          pip install --user grpcio-tools==1.64.0
+          
       - name: Test sentencetransformers
         run: |
            make --jobs=5 --output-sync=target -C backend/python/sentencetransformers
@@ -64,7 +64,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -74,7 +74,7 @@ jobs:
           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
+          pip install --user grpcio-tools==1.64.0
 
       - name: Test rerankers
         run: |
@@ -86,7 +86,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -96,7 +96,7 @@ jobs:
           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
+          pip install --user grpcio-tools==1.64.0
       - name: Test diffusers
         run: |
           make --jobs=5 --output-sync=target -C backend/python/diffusers
@@ -107,7 +107,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -117,19 +117,19 @@ jobs:
           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
+          pip install --user grpcio-tools==1.64.0
 
       - name: Test parler-tts
         run: |
            make --jobs=5 --output-sync=target -C backend/python/parler-tts
            make --jobs=5 --output-sync=target -C backend/python/parler-tts test
-
+  
   tests-openvoice:
     runs-on: ubuntu-latest
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -139,7 +139,7 @@ jobs:
           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
+          pip install --user grpcio-tools==1.64.0
 
       - name: Test openvoice
         run: |
@@ -151,7 +151,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -161,7 +161,7 @@ jobs:
           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
+          pip install --user grpcio-tools==1.64.0
 
       - name: Test transformers-musicgen
         run: |
@@ -175,7 +175,7 @@ jobs:
   #   steps:
   #     - name: Clone
   #       uses: actions/checkout@v4
-  #       with:
+  #       with: 
   #         submodules: true
   #     - name: Dependencies
   #       run: |
@@ -185,14 +185,14 @@ jobs:
   #         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
+  #         pip install --user grpcio-tools==1.64.0
 
   #     - name: Test petals
   #       run: |
   #          make --jobs=5 --output-sync=target -C backend/python/petals
   #          make --jobs=5 --output-sync=target -C backend/python/petals test
 
-
+           
 
   # tests-bark:
   #   runs-on: ubuntu-latest
@@ -239,7 +239,7 @@ jobs:
   #           df -h
   #     - name: Clone
   #       uses: actions/checkout@v4
-  #       with:
+  #       with: 
   #         submodules: true
   #     - name: Dependencies
   #       run: |
@@ -249,14 +249,14 @@ jobs:
   #         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
+  #         pip install --user grpcio-tools==1.64.0
 
   #     - 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:
@@ -264,7 +264,7 @@ jobs:
   #   steps:
   #     - name: Clone
   #       uses: actions/checkout@v4
-  #       with:
+  #       with: 
   #         submodules: true
   #     - name: Dependencies
   #       run: |
@@ -274,7 +274,7 @@ jobs:
   #         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
+  #         pip install --user grpcio-tools==1.64.0
   #     - name: Test vllm
   #       run: |
   #          make --jobs=5 --output-sync=target -C backend/python/vllm
@@ -284,7 +284,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -294,7 +294,7 @@ jobs:
           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
+          pip install --user grpcio-tools==1.64.0
       - name: Test vall-e-x
         run: |
            make --jobs=5 --output-sync=target -C backend/python/vall-e-x
@@ -305,7 +305,7 @@ jobs:
     steps:
       - name: Clone
         uses: actions/checkout@v4
-        with:
+        with: 
           submodules: true
       - name: Dependencies
         run: |
@@ -314,8 +314,8 @@ jobs:
           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
+          pip install --user grpcio-tools==1.64.0
       - 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
+          make --jobs=5 --output-sync=target -C backend/python/coqui test

.github/workflows/test.yml

@@ -10,7 +10,7 @@ on:
       - '*'
 
 env:
-  GRPC_VERSION: v1.65.0
+  GRPC_VERSION: v1.64.0
 
 concurrency:
   group: ci-tests-${{ github.head_ref || github.ref }}-${{ github.repository }}
@@ -71,7 +71,6 @@ jobs:
         run: |
           sudo apt-get update
           sudo apt-get install build-essential 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 && \
@@ -94,8 +93,8 @@ jobs:
           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
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
 
           # The python3-grpc-tools package in 22.04 is too old
           pip install --user grpcio-tools
@@ -110,7 +109,7 @@ jobs:
           # Pre-build stable diffusion before we install a newer version of abseil (not compatible with stablediffusion-ncn)
           PATH="$PATH:/root/go/bin" GO_TAGS="stablediffusion tts" GRPC_BACKENDS=backend-assets/grpc/stablediffusion make build
         env:
-          CUDA_VERSION: 12-4
+          CUDA_VERSION: 12-3
       - name: Cache grpc
         id: cache-grpc
         uses: actions/cache@v4
@@ -121,8 +120,7 @@ jobs:
         if: steps.cache-grpc.outputs.cache-hit != 'true'
         run: |
           git clone --recurse-submodules -b ${{ env.GRPC_VERSION }} --depth 1 --jobs 5 --shallow-submodules https://github.com/grpc/grpc && \
-          cd grpc && sed -i "216i\  TESTONLY" "third_party/abseil-cpp/absl/container/CMakeLists.txt" && mkdir -p cmake/build && cd cmake/build && \
-          cmake -DgRPC_INSTALL=ON \
+          cd grpc && mkdir -p cmake/build && cd cmake/build && cmake -DgRPC_INSTALL=ON \
             -DgRPC_BUILD_TESTS=OFF \
             ../.. && sudo make --jobs 5
       - name: Install gRPC
@@ -215,14 +213,14 @@ jobs:
       - name: Dependencies
         run: |
           brew install protobuf grpc make protoc-gen-go protoc-gen-go-grpc
-          pip install --user --no-cache-dir grpcio-tools==1.64.1
+          pip install --user grpcio-tools==1.64.0
       - name: Test
         run: |
           export C_INCLUDE_PATH=/usr/local/include
           export CPLUS_INCLUDE_PATH=/usr/local/include
           # Used to run the newer GNUMake version from brew that supports --output-sync
           export PATH="/opt/homebrew/opt/make/libexec/gnubin:$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
+          BUILD_TYPE="GITHUB_CI_HAS_BROKEN_METAL" CMAKE_ARGS="-DLLAMA_F16C=OFF -DLLAMA_AVX512=OFF -DLLAMA_AVX2=OFF -DLLAMA_FMA=OFF" make --jobs 4 --output-sync=target test
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
         uses: mxschmitt/action-tmate@v3.18

.github/workflows/update_swagger.yaml

@@ -13,17 +13,11 @@ jobs:
       - 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
+          make swagger
       - name: Create Pull Request
         uses: peter-evans/create-pull-request@v6
         with:

Dockerfile

@@ -8,7 +8,7 @@ FROM ${BASE_IMAGE} AS requirements-core
 
 USER root
 
-ARG GO_VERSION=1.22.5
+ARG GO_VERSION=1.22.4
 ARG TARGETARCH
 ARG TARGETVARIANT
 
@@ -99,7 +99,7 @@ FROM requirements-${IMAGE_TYPE} AS requirements-drivers
 
 ARG BUILD_TYPE
 ARG CUDA_MAJOR_VERSION=12
-ARG CUDA_MINOR_VERSION=4
+ARG CUDA_MINOR_VERSION=5
 
 ENV BUILD_TYPE=${BUILD_TYPE}
 
@@ -108,11 +108,11 @@ RUN <<EOT bash
     if [ "${BUILD_TYPE}" = "vulkan" ]; then
         apt-get update && \
         apt-get install -y  --no-install-recommends \
-            software-properties-common pciutils wget gpg-agent && \
+                        software-properties-common pciutils wget gpg-agent && \
         wget -qO - https://packages.lunarg.com/lunarg-signing-key-pub.asc | apt-key add - && \
         wget -qO /etc/apt/sources.list.d/lunarg-vulkan-jammy.list https://packages.lunarg.com/vulkan/lunarg-vulkan-jammy.list && \
         apt-get update && \
-        apt-get install -y \
+            apt-get install -y \
             vulkan-sdk && \
         apt-get clean && \
         rm -rf /var/lib/apt/lists/*
@@ -124,13 +124,33 @@ RUN <<EOT bash
     if [ "${BUILD_TYPE}" = "cublas" ]; then
         apt-get update && \
         apt-get install -y  --no-install-recommends \
-            software-properties-common pciutils
+                        software-properties-common pciutils
         if [ "amd64" = "$TARGETARCH" ]; then
             curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
-        fi
+            fi
         if [ "arm64" = "$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
+
+RUN if [ "${BUILD_TYPE}" = "cublas" ]; then \
+        apt-get update && \
+        apt-get install -y  --no-install-recommends \
+            software-properties-common pciutils && \
+        curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb && \
         dpkg -i cuda-keyring_1.1-1_all.deb && \
         rm -f cuda-keyring_1.1-1_all.deb && \
         apt-get update && \
@@ -142,9 +162,8 @@ RUN <<EOT bash
             libcusparse-dev-${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
+        rm -rf /var/lib/apt/lists/* \
+    ; fi
 
 # If we are building with clblas support, we need the libraries for the builds
 RUN if [ "${BUILD_TYPE}" = "clblas" ]; then \
@@ -187,7 +206,7 @@ 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 GRPC_VERSION=v1.64.2
 
 ENV MAKEFLAGS=${GRPC_MAKEFLAGS}
 
@@ -208,7 +227,6 @@ RUN apt-get update && \
 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 && \
@@ -264,8 +282,6 @@ COPY --from=grpc /opt/grpc /usr/local
 
 # Rebuild with defaults backends
 WORKDIR /build
-
-## Build the binary
 RUN make build
 
 RUN if [ ! -d "/build/sources/go-piper/piper-phonemize/pi/lib/" ]; then \

Makefile

@@ -3,12 +3,9 @@ GOTEST=$(GOCMD) test
 GOVET=$(GOCMD) vet
 BINARY_NAME=local-ai
 
-DETECT_LIBS?=true
-
 # llama.cpp versions
-GOLLAMA_REPO?=https://github.com/go-skynet/go-llama.cpp
-GOLLAMA_VERSION?=2b57a8ae43e4699d3dc5d1496a1ccd42922993be
-CPPLLAMA_VERSION?=87e397d00bdcedd5cbf6dfda06a7b0f302462728
+GOLLAMA_STABLE_VERSION?=2b57a8ae43e4699d3dc5d1496a1ccd42922993be
+CPPLLAMA_VERSION?=e112b610a1a75cb7fa8351e1a933e2e7a755a5ce
 
 # gpt4all version
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
@@ -19,33 +16,26 @@ RWKV_REPO?=https://github.com/donomii/go-rwkv.cpp
 RWKV_VERSION?=661e7ae26d442f5cfebd2a0881b44e8c55949ec6
 
 # whisper.cpp version
-WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp
-WHISPER_CPP_VERSION?=f68298ce06ca3edd6e6f3f21c3d0bb5f073942c3
+WHISPER_CPP_VERSION?=b29b3b29240aac8b71ce8e5a4360c1f1562ad66f
 
 # bert.cpp version
-BERT_REPO?=https://github.com/go-skynet/go-bert.cpp
 BERT_VERSION?=710044b124545415f555e4260d16b146c725a6e4
 
 # go-piper version
-PIPER_REPO?=https://github.com/mudler/go-piper
 PIPER_VERSION?=9d0100873a7dbb0824dfea40e8cec70a1b110759
 
 # stablediffusion version
-STABLEDIFFUSION_REPO?=https://github.com/mudler/go-stable-diffusion
 STABLEDIFFUSION_VERSION?=4a3cd6aeae6f66ee57eae9a0075f8c58c3a6a38f
 
 # tinydream version
-TINYDREAM_REPO?=https://github.com/M0Rf30/go-tiny-dream
 TINYDREAM_VERSION?=c04fa463ace9d9a6464313aa5f9cd0f953b6c057
 
 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?=
@@ -59,12 +49,12 @@ RANDOM := $(shell bash -c 'echo $$RANDOM')
 VERSION?=$(shell git describe --always --tags || echo "dev" )
 # go tool nm ./local-ai | grep Commit
 LD_FLAGS?=
-override LD_FLAGS += -X "github.com/mudler/LocalAI/internal.Version=$(VERSION)"
-override LD_FLAGS += -X "github.com/mudler/LocalAI/internal.Commit=$(shell git rev-parse HEAD)"
+override LD_FLAGS += -X "github.com/go-skynet/LocalAI/internal.Version=$(VERSION)"
+override LD_FLAGS += -X "github.com/go-skynet/LocalAI/internal.Commit=$(shell git rev-parse HEAD)"
 
 OPTIONAL_TARGETS?=
 
-export OS := $(shell uname -s)
+OS := $(shell uname -s)
 ARCH := $(shell uname -m)
 GREEN  := $(shell tput -Txterm setaf 2)
 YELLOW := $(shell tput -Txterm setaf 3)
@@ -90,40 +80,31 @@ ifeq ($(OS),Darwin)
 		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
+		CMAKE_ARGS+=-DLLAMA_METAL=OFF
+		export LLAMA_NO_ACCELERATE=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
+	export WHISPER_OPENBLAS=1
 endif
 
+
 ifeq ($(BUILD_TYPE),cublas)
 	CGO_LDFLAGS+=-lcublas -lcudart -L$(CUDA_LIBPATH)
-	export GGML_CUDA=1
+	export LLAMA_CUBLAS=1
+	export WHISPER_CUDA=1
 	CGO_LDFLAGS_WHISPER+=-L$(CUDA_LIBPATH)/stubs/ -lcuda -lcufft
 endif
 
 ifeq ($(BUILD_TYPE),vulkan)
-	CMAKE_ARGS+=-DGGML_VULKAN=1
-endif
-
-ifneq (,$(findstring sycl,$(BUILD_TYPE)))
-	export GGML_SYCL=1
-endif
-
-ifeq ($(BUILD_TYPE),sycl_f16)
-	export GGML_SYCL_F16=1
+	CMAKE_ARGS+=-DLLAMA_VULKAN=1
 endif
 
 ifeq ($(BUILD_TYPE),hipblas)
@@ -134,26 +115,27 @@ ifeq ($(BUILD_TYPE),hipblas)
 	export CC=$(ROCM_HOME)/llvm/bin/clang
 	# llama-ggml has no hipblas support, so override it here.
 	export STABLE_BUILD_TYPE=
-	export GGML_HIPBLAS=1
+	export WHISPER_HIPBLAS=1
 	GPU_TARGETS ?= gfx900,gfx906,gfx908,gfx940,gfx941,gfx942,gfx90a,gfx1030,gfx1031,gfx1100,gfx1101
 	AMDGPU_TARGETS ?= "$(GPU_TARGETS)"
-	CMAKE_ARGS+=-DGGML_HIPBLAS=ON -DAMDGPU_TARGETS="$(AMDGPU_TARGETS)" -DGPU_TARGETS="$(GPU_TARGETS)"
+	CMAKE_ARGS+=-DLLAMA_HIPBLAS=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
+	export LLAMA_METAL=1
+	export WHISPER_METAL=1
 endif
 
 ifeq ($(BUILD_TYPE),clblas)
 	CGO_LDFLAGS+=-lOpenCL -lclblast
-	export GGML_OPENBLAS=1
+	export WHISPER_CLBLAST=1
 endif
 
 # glibc-static or glibc-devel-static required
 ifeq ($(STATIC),true)
-	LD_FLAGS+=-linkmode external -extldflags -static
+	LD_FLAGS=-linkmode external -extldflags -static
 endif
 
 ifeq ($(findstring stablediffusion,$(GO_TAGS)),stablediffusion)
@@ -187,8 +169,6 @@ ALL_GRPC_BACKENDS+=backend-assets/grpc/rwkv
 ALL_GRPC_BACKENDS+=backend-assets/grpc/whisper
 ALL_GRPC_BACKENDS+=backend-assets/grpc/local-store
 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/...
@@ -208,109 +188,69 @@ all: help
 
 ## BERT embeddings
 sources/go-bert.cpp:
-	mkdir -p sources/go-bert.cpp
-	cd sources/go-bert.cpp && \
-	git init && \
-	git remote add origin $(BERT_REPO) && \
-	git fetch origin && \
-	git checkout $(BERT_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
+	git clone --recurse-submodules https://github.com/go-skynet/go-bert.cpp sources/go-bert.cpp
+	cd sources/go-bert.cpp && git checkout -b build $(BERT_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/go-bert.cpp/libgobert.a: sources/go-bert.cpp
 	$(MAKE) -C sources/go-bert.cpp libgobert.a
 
 ## go-llama.cpp
 sources/go-llama.cpp:
-	mkdir -p sources/go-llama.cpp
-	cd sources/go-llama.cpp && \
-	git init && \
-	git remote add origin $(GOLLAMA_REPO) && \
-	git fetch origin && \
-	git checkout $(GOLLAMA_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
+	git clone --recurse-submodules https://github.com/go-skynet/go-llama.cpp sources/go-llama.cpp
+	cd sources/go-llama.cpp && git checkout -b build $(GOLLAMA_STABLE_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/go-llama.cpp/libbinding.a: sources/go-llama.cpp
 	$(MAKE) -C sources/go-llama.cpp BUILD_TYPE=$(STABLE_BUILD_TYPE) libbinding.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
+	git clone --recurse-submodules https://github.com/mudler/go-piper sources/go-piper
+	cd sources/go-piper && git checkout -b build $(PIPER_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/go-piper/libpiper_binding.a: sources/go-piper
 	$(MAKE) -C sources/go-piper libpiper_binding.a example/main piper.o
 
 ## GPT4ALL
 sources/gpt4all:
-	mkdir -p sources/gpt4all
-	cd sources/gpt4all && \
-	git init && \
-	git remote add origin $(GPT4ALL_REPO) && \
-	git fetch origin && \
-	git checkout $(GPT4ALL_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
+	git clone --recurse-submodules $(GPT4ALL_REPO) sources/gpt4all
+	cd sources/gpt4all && git checkout -b build $(GPT4ALL_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/gpt4all/gpt4all-bindings/golang/libgpt4all.a: sources/gpt4all
 	$(MAKE) -C sources/gpt4all/gpt4all-bindings/golang/ libgpt4all.a
 
 ## RWKV
 sources/go-rwkv.cpp:
-	mkdir -p sources/go-rwkv.cpp
-	cd sources/go-rwkv.cpp && \
-	git init && \
-	git remote add origin $(RWKV_REPO) && \
-	git fetch origin && \
-	git checkout $(RWKV_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
+	git clone --recurse-submodules $(RWKV_REPO) sources/go-rwkv.cpp
+	cd sources/go-rwkv.cpp && git checkout -b build $(RWKV_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/go-rwkv.cpp/librwkv.a: sources/go-rwkv.cpp
 	cd sources/go-rwkv.cpp && cd rwkv.cpp &&	cmake . -DRWKV_BUILD_SHARED_LIBRARY=OFF &&	cmake --build . && 	cp librwkv.a ..
 
 ## stable diffusion
 sources/go-stable-diffusion:
-	mkdir -p sources/go-stable-diffusion
-	cd sources/go-stable-diffusion && \
-	git init && \
-	git remote add origin $(STABLEDIFFUSION_REPO) && \
-	git fetch origin && \
-	git checkout $(STABLEDIFFUSION_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
+	git clone --recurse-submodules https://github.com/mudler/go-stable-diffusion sources/go-stable-diffusion
+	cd sources/go-stable-diffusion && git checkout -b build $(STABLEDIFFUSION_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/go-stable-diffusion/libstablediffusion.a: sources/go-stable-diffusion
 	CPATH="$(CPATH):/usr/include/opencv4" $(MAKE) -C sources/go-stable-diffusion libstablediffusion.a
 
 ## tiny-dream
 sources/go-tiny-dream:
-	mkdir -p sources/go-tiny-dream
-	cd sources/go-tiny-dream && \
-	git init && \
-	git remote add origin $(TINYDREAM_REPO) && \
-	git fetch origin && \
-	git checkout $(TINYDREAM_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
+	git clone --recurse-submodules https://github.com/M0Rf30/go-tiny-dream sources/go-tiny-dream
+	cd sources/go-tiny-dream && git checkout -b build $(TINYDREAM_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/go-tiny-dream/libtinydream.a: sources/go-tiny-dream
 	$(MAKE) -C sources/go-tiny-dream libtinydream.a
 
 ## 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
+	git clone https://github.com/ggerganov/whisper.cpp sources/whisper.cpp
+	cd sources/whisper.cpp && git checkout -b build $(WHISPER_CPP_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/whisper.cpp/libwhisper.a: sources/whisper.cpp
-	cd sources/whisper.cpp && $(MAKE) libwhisper.a libggml.a
+	cd sources/whisper.cpp && $(MAKE) libwhisper.a
 
-get-sources: sources/go-llama.cpp sources/gpt4all sources/go-piper sources/go-rwkv.cpp sources/whisper.cpp sources/go-bert.cpp sources/go-stable-diffusion sources/go-tiny-dream backend/cpp/llama/llama.cpp
+get-sources: sources/go-llama.cpp sources/gpt4all sources/go-piper sources/go-rwkv.cpp sources/whisper.cpp sources/go-bert.cpp sources/go-stable-diffusion sources/go-tiny-dream
 
 replace:
 	$(GOCMD) mod edit -replace github.com/donomii/go-rwkv.cpp=$(CURDIR)/sources/go-rwkv.cpp
@@ -379,12 +319,12 @@ build: prepare backend-assets grpcs ## Build the project
 	$(info ${GREEN}I LD_FLAGS: ${YELLOW}$(LD_FLAGS)${RESET})
 ifneq ($(BACKEND_LIBS),)
 	$(MAKE) backend-assets/lib
-	cp -f $(BACKEND_LIBS) backend-assets/lib/
+	cp $(BACKEND_LIBS) backend-assets/lib/
 endif
 	CGO_LDFLAGS="$(CGO_LDFLAGS)" $(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o $(BINARY_NAME) ./
 
 build-minimal:
-	BUILD_GRPC_FOR_BACKEND_LLAMA=true GRPC_BACKENDS="backend-assets/grpc/llama-cpp-avx2" GO_TAGS=p2p $(MAKE) build
+	BUILD_GRPC_FOR_BACKEND_LLAMA=true GRPC_BACKENDS="backend-assets/grpc/llama-cpp-avx2" GO_TAGS=none $(MAKE) build
 
 build-api:
 	BUILD_GRPC_FOR_BACKEND_LLAMA=true BUILD_API_ONLY=true GO_TAGS=none $(MAKE) build
@@ -394,9 +334,6 @@ 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)
 	$(info ${GREEN}I Skip CUDA/hipblas build on MacOS${RESET})
 else
@@ -405,11 +342,7 @@ else
 	$(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
+	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),)
@@ -420,8 +353,8 @@ else
 	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="$(CMAKE_ARGS) -DGGML_NATIVE=off" GRPC_BACKENDS="backend-assets/grpc/llama-cpp-fallback backend-assets/grpc/llama-cpp-grpc backend-assets/util/llama-cpp-rpc-server" \
+dist-cross-linux-arm64: 
+	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_NATIVE=off" GRPC_BACKENDS="backend-assets/grpc/llama-cpp-fallback backend-assets/grpc/llama-cpp-grpc backend-assets/util/llama-cpp-rpc-server" \
 	STATIC=true $(MAKE) build
 	mkdir -p release
 # if BUILD_ID is empty, then we don't append it to the binary name
@@ -471,7 +404,7 @@ 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=$(BUILD_TYPE) --build-arg CUDA_MAJOR_VERSION=12 --build-arg CUDA_MINOR_VERSION=4 --build-arg FFMPEG=true -t localai-tests .
+	docker build --build-arg GRPC_BACKENDS="$(GRPC_BACKENDS)" --build-arg IMAGE_TYPE=core --build-arg BUILD_TYPE=$(BUILD_TYPE) --build-arg CUDA_MAJOR_VERSION=12 --build-arg CUDA_MINOR_VERSION=5 --build-arg FFMPEG=true -t localai-tests .
 
 run-e2e-image:
 	ls -liah $(abspath ./tests/e2e-fixtures)
@@ -767,71 +700,71 @@ else
 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
+backend-assets/grpc/llama-cpp: backend-assets/grpc
 	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
+backend-assets/grpc/llama-cpp-avx2: backend-assets/grpc
 	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" $(MAKE) VARIANT="llama-avx2" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=on -DLLAMA_AVX2=on -DLLAMA_AVX512=off -DLLAMA_FMA=on -DLLAMA_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-avx: backend-assets/grpc backend/cpp/llama/llama.cpp
+backend-assets/grpc/llama-cpp-avx: backend-assets/grpc
 	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" $(MAKE) VARIANT="llama-avx" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=on -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_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
+backend-assets/grpc/llama-cpp-fallback: backend-assets/grpc
 	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" $(MAKE) VARIANT="llama-fallback" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=off -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off" $(MAKE) VARIANT="llama-fallback" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-fallback/grpc-server backend-assets/grpc/llama-cpp-fallback
 # TODO: every binary should have its own folder instead, so can have different metal implementations
 ifeq ($(BUILD_TYPE),metal)
 	cp backend/cpp/llama-fallback/llama.cpp/build/bin/default.metallib backend-assets/grpc/
 endif
 
-backend-assets/grpc/llama-cpp-cuda: backend-assets/grpc backend/cpp/llama/llama.cpp
+backend-assets/grpc/llama-cpp-cuda: backend-assets/grpc
 	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" $(MAKE) VARIANT="llama-cuda" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=on -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off -DLLAMA_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
+backend-assets/grpc/llama-cpp-hipblas: backend-assets/grpc
 	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})
 	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
+backend-assets/grpc/llama-cpp-sycl_f16: backend-assets/grpc
 	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
+backend-assets/grpc/llama-cpp-sycl_f32: backend-assets/grpc
 	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
+backend-assets/grpc/llama-cpp-grpc: backend-assets/grpc
 	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" $(MAKE) VARIANT="llama-grpc" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_RPC=ON -DLLAMA_AVX=off -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off" $(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
@@ -859,7 +792,7 @@ backend-assets/grpc/tinydream: sources/go-tiny-dream sources/go-tiny-dream/libti
 	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/tinydream ./backend/go/image/tinydream
 
 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 \
+	CGO_LDFLAGS="$(CGO_LDFLAGS) $(CGO_LDFLAGS_WHISPER)" C_INCLUDE_PATH=$(CURDIR)/sources/whisper.cpp LIBRARY_PATH=$(CURDIR)/sources/whisper.cpp \
 	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/whisper ./backend/go/transcribe/
 
 backend-assets/grpc/local-store: backend-assets/grpc
@@ -905,7 +838,7 @@ docker-aio-all:
 
 docker-image-intel:
 	docker build \
-		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.2.0-devel-ubuntu22.04 \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.1.0-devel-ubuntu22.04 \
 		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
 		--build-arg GO_TAGS="none" \
 		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
@@ -913,7 +846,7 @@ docker-image-intel:
 
 docker-image-intel-xpu:
 	docker build \
-		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.2.0-devel-ubuntu22.04 \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.1.0-devel-ubuntu22.04 \
 		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
 		--build-arg GO_TAGS="none" \
 		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
@@ -928,7 +861,7 @@ gen-assets:
 	$(GOCMD) run core/dependencies_manager/manager.go embedded/webui_static.yaml core/http/static/assets
 
 ## Documentation
-docs/layouts/_default:
+docs/layouts/_default: 
 	mkdir -p docs/layouts/_default
 
 docs/static/gallery.html: docs/layouts/_default
@@ -943,4 +876,4 @@ docs-clean:
 
 .PHONY: docs
 docs: docs/static/gallery.html
-	cd docs && hugo serve
+	cd docs && hugo serve

README.md

@@ -72,15 +72,13 @@ docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-aio-cpu
 
 [Roadmap](https://github.com/mudler/LocalAI/issues?q=is%3Aissue+is%3Aopen+label%3Aroadmap)
 
-- July 2024: 🔥🔥 🆕 P2P Dashboard, LocalAI Federated mode and AI Swarms: https://github.com/mudler/LocalAI/pull/2723
-- June 2024: 🆕 You can browse now the model gallery without LocalAI! Check out https://models.localai.io
-- June 2024: Support for models from OCI registries: https://github.com/mudler/LocalAI/pull/2628
-- May 2024: 🔥🔥 Decentralized P2P llama.cpp:  https://github.com/mudler/LocalAI/pull/2343 (peer2peer llama.cpp!) 👉 Docs  https://localai.io/features/distribute/
-- May 2024: 🔥🔥 Openvoice: https://github.com/mudler/LocalAI/pull/2334
-- May 2024: 🆕 Function calls without grammars and mixed mode: https://github.com/mudler/LocalAI/pull/2328
-- May 2024: 🔥🔥 Distributed inferencing: https://github.com/mudler/LocalAI/pull/2324
-- May 2024: Chat, TTS, and Image generation in the WebUI: https://github.com/mudler/LocalAI/pull/2222
-- April 2024: Reranker API: https://github.com/mudler/LocalAI/pull/2121
+- 🆕 You can browse now the model gallery without LocalAI! Check out https://models.localai.io
+- 🔥🔥 Decentralized llama.cpp:  https://github.com/mudler/LocalAI/pull/2343 (peer2peer llama.cpp!) 👉 Docs  https://localai.io/features/distribute/
+- 🔥🔥 Openvoice: https://github.com/mudler/LocalAI/pull/2334
+- 🆕 Function calls without grammars and mixed mode: https://github.com/mudler/LocalAI/pull/2328
+- 🔥🔥 Distributed inferencing: https://github.com/mudler/LocalAI/pull/2324
+- Chat, TTS, and Image generation in the WebUI: https://github.com/mudler/LocalAI/pull/2222
+- Reranker API: https://github.com/mudler/LocalAI/pull/2121
 
 Hot topics (looking for contributors):
 
@@ -90,7 +88,6 @@ Hot topics (looking for contributors):
 - Assistant API: https://github.com/mudler/LocalAI/issues/1273
 - Moderation endpoint: https://github.com/mudler/LocalAI/issues/999
 - Vulkan: https://github.com/mudler/LocalAI/issues/1647
-- Anthropic API: https://github.com/mudler/LocalAI/issues/1808
 
 If you want to help and contribute, issues up for grabs: https://github.com/mudler/LocalAI/issues?q=is%3Aissue+is%3Aopen+label%3A%22up+for+grabs%22
 
@@ -107,7 +104,6 @@ If you want to help and contribute, issues up for grabs: https://github.com/mudl
 - 🥽 [Vision API](https://localai.io/features/gpt-vision/)
 - 📈 [Reranker API](https://localai.io/features/reranker/)
 - 🆕🖧 [P2P Inferencing](https://localai.io/features/distribute/)
-- 🌍 Integrated WebUI!
 
 ## 💻 Usage
 
@@ -136,7 +132,6 @@ Other:
 - Slack bot https://github.com/mudler/LocalAGI/tree/main/examples/slack
 - Shell-Pilot(Interact with LLM using LocalAI models via pure shell scripts on your Linux or MacOS system) https://github.com/reid41/shell-pilot
 - Telegram bot https://github.com/mudler/LocalAI/tree/master/examples/telegram-bot
-- Github Actions: https://github.com/marketplace/actions/start-localai
 - Examples: https://github.com/mudler/LocalAI/tree/master/examples/
   
 

backend/cpp/grpc/Makefile

@@ -46,14 +46,9 @@ endif
 $(INSTALLED_PACKAGES): grpc_build
 
 $(GRPC_REPO):
-	mkdir -p $(GRPC_REPO)/grpc
-	cd $(GRPC_REPO)/grpc && \
-	git init && \
-	git remote add origin $(GIT_REPO_LIB_GRPC)  && \
-	git fetch origin && \
-	git checkout $(TAG_LIB_GRPC) && \
-	git submodule update --init --recursive --depth 1 --single-branch
-	
+	git clone --depth $(GIT_CLONE_DEPTH) -b $(TAG_LIB_GRPC) $(GIT_REPO_LIB_GRPC) $(GRPC_REPO)/grpc
+	cd $(GRPC_REPO)/grpc && git submodule update --jobs 2 --init --recursive --depth $(GIT_CLONE_DEPTH)
+
 $(GRPC_BUILD): $(GRPC_REPO)
 	mkdir -p $(GRPC_BUILD)
 	cd $(GRPC_BUILD) && cmake $(CMAKE_ARGS) ../$(GRPC_REPO)/grpc && cmake --build . && cmake --build . --target install

backend/cpp/llama/Makefile

@@ -1,58 +1,45 @@
 
 LLAMA_VERSION?=
-LLAMA_REPO?=https://github.com/ggerganov/llama.cpp
 
 CMAKE_ARGS?=
 BUILD_TYPE?=
 ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh
-TARGET?=--target grpc-server
 
-# Disable Shared libs as we are linking on static gRPC and we can't mix shared and static
-CMAKE_ARGS+=-DBUILD_SHARED_LIBS=OFF
-
-# If build type is cublas, then we set -DGGML_CUDA=ON to CMAKE_ARGS automatically
+# If build type is cublas, then we set -DLLAMA_CUBLAS=ON to CMAKE_ARGS automatically
 ifeq ($(BUILD_TYPE),cublas)
-	CMAKE_ARGS+=-DGGML_CUDA=ON
-# If build type is openblas then we set -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
+	CMAKE_ARGS+=-DLLAMA_CUBLAS=ON
+# If build type is openblas then we set -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
 # to CMAKE_ARGS automatically
 else ifeq ($(BUILD_TYPE),openblas)
-	CMAKE_ARGS+=-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
-# If build type is clblas (openCL) we set -DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
+	CMAKE_ARGS+=-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
+# If build type is clblas (openCL) we set -DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
 else ifeq ($(BUILD_TYPE),clblas)
-	CMAKE_ARGS+=-DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
+	CMAKE_ARGS+=-DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
 # If it's hipblas we do have also to set CC=/opt/rocm/llvm/bin/clang CXX=/opt/rocm/llvm/bin/clang++ 
 else ifeq ($(BUILD_TYPE),hipblas)
-	CMAKE_ARGS+=-DGGML_HIPBLAS=ON
-# If it's OSX, DO NOT embed the metal library - -DGGML_METAL_EMBED_LIBRARY=ON requires further investigation
+	CMAKE_ARGS+=-DLLAMA_HIPBLAS=ON
+# If it's OSX, DO NOT embed the metal library - -DLLAMA_METAL_EMBED_LIBRARY=ON requires further investigation
 # But if it's OSX without metal, disable it here
-else ifeq ($(OS),Darwin)
+else ifeq ($(OS),darwin)
 	ifneq ($(BUILD_TYPE),metal)
-		CMAKE_ARGS+=-DGGML_METAL=OFF
-	else
-		CMAKE_ARGS+=-DGGML_METAL=ON
-# Until this is tested properly, we disable embedded metal file
-# as we already embed it as part of the LocalAI assets
-		CMAKE_ARGS+=-DGGML_METAL_EMBED_LIBRARY=OFF
-		TARGET+=--target ggml-metal
+		CMAKE_ARGS+=-DLLAMA_METAL=OFF
 	endif
 endif
 
 ifeq ($(BUILD_TYPE),sycl_f16)
-	CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DGGML_SYCL_F16=ON
+	CMAKE_ARGS+=-DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
 endif
 
 ifeq ($(BUILD_TYPE),sycl_f32)
-	CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
+	CMAKE_ARGS+=-DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
 endif
 
 llama.cpp:
-	mkdir -p llama.cpp
-	cd llama.cpp && \
-	git init && \
-	git remote add origin $(LLAMA_REPO)  && \
-	git fetch origin && \
-	git checkout -b build $(LLAMA_VERSION) && \
-	git submodule update --init --recursive --depth 1 --single-branch
+	git clone --recurse-submodules https://github.com/ggerganov/llama.cpp llama.cpp
+	if [ -z "$(LLAMA_VERSION)" ]; then \
+		exit 1; \
+	fi
+	cd llama.cpp && git checkout -b build $(LLAMA_VERSION) && git submodule update --init --recursive --depth 1
 
 llama.cpp/examples/grpc-server: llama.cpp
 	mkdir -p llama.cpp/examples/grpc-server
@@ -74,9 +61,9 @@ clean: purge
 grpc-server: llama.cpp llama.cpp/examples/grpc-server
 	@echo "Building grpc-server with $(BUILD_TYPE) build type and $(CMAKE_ARGS)"
 ifneq (,$(findstring sycl,$(BUILD_TYPE)))
-	+bash -c "source $(ONEAPI_VARS); \
-	cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && cmake --build . --config Release $(TARGET)"
+	bash -c "source $(ONEAPI_VARS); \
+	cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && $(MAKE)"
 else
-	+cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && cmake --build . --config Release $(TARGET)
+	cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && $(MAKE)
 endif
 	cp llama.cpp/build/bin/grpc-server .

backend/cpp/llama/grpc-server.cpp

@@ -886,8 +886,6 @@ struct llama_server_context
             {"task_id", slot->task_id},
         });
 
-        LOG_TEE("sampling: \n%s\n", llama_sampling_print(slot->sparams).c_str());
-
         return true;
     }
 
@@ -2108,7 +2106,6 @@ json parse_options(bool streaming, const backend::PredictOptions* predict, llama
     data["grammar"] = predict->grammar();
     data["prompt"] = predict->prompt();
     data["ignore_eos"] = predict->ignoreeos();
-    data["embeddings"] = predict->embeddings();
 
     // for each image in the request, add the image data
     //
@@ -2386,31 +2383,6 @@ public:
 
         return grpc::Status::OK;
     }
-
-    /// https://github.com/ggerganov/llama.cpp/blob/aa2341298924ac89778252015efcb792f2df1e20/examples/server/server.cpp#L2969
-    grpc::Status Embedding(ServerContext* context, const backend::PredictOptions* request, backend::EmbeddingResult* embeddingResult) {
-        json data = parse_options(false, request, llama);
-        const int task_id = llama.queue_tasks.get_new_id();
-        llama.queue_results.add_waiting_task_id(task_id);
-        llama.request_completion(task_id, { {"prompt", data["embeddings"]}, { "n_predict", 0}, {"image_data", ""} }, false, true, -1);
-        // get the result
-        task_result result = llama.queue_results.recv(task_id);
-        //std::cout << "Embedding result JSON" << result.result_json.dump() << std::endl;
-        llama.queue_results.remove_waiting_task_id(task_id);
-        if (!result.error && result.stop) {
-            std::vector<float> embeddings = result.result_json.value("embedding", std::vector<float>());
-            // loop the vector and set the embeddings results
-            for (int i = 0; i < embeddings.size(); i++) {
-                embeddingResult->add_embeddings(embeddings[i]);
-            }
-        }
-        else
-        {
-            return grpc::Status::OK;
-        }
-
-        return grpc::Status::OK;
-    }
 };
 
 void RunServer(const std::string& server_address) {

backend/go/llm/llama/llama.go

@@ -6,9 +6,9 @@ import (
 	"fmt"
 	"path/filepath"
 
+	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
 	"github.com/go-skynet/go-llama.cpp"
 	"github.com/mudler/LocalAI/pkg/grpc/base"
-	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 )
 
 type LLM struct {

backend/python/autogptq/requirements-intel.txt

@@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/autogptq/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 auto-gptq==0.7.1
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 torch
 certifi

backend/python/bark/requirements-intel.txt

@@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/bark/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 bark==0.1.5
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 certifi
 transformers

backend/python/common/libbackend.sh

@@ -148,13 +148,13 @@ function startBackend() {
     ensureVenv
 
     if [ ! -z ${BACKEND_FILE} ]; then
-        exec python ${BACKEND_FILE} $@
+        python ${BACKEND_FILE} $@
     elif [ -e "${MY_DIR}/server.py" ]; then
-        exec python ${MY_DIR}/server.py $@
+        python ${MY_DIR}/server.py $@
     elif [ -e "${MY_DIR}/backend.py" ]; then
-        exec python ${MY_DIR}/backend.py $@
+        python ${MY_DIR}/backend.py $@
     elif [ -e "${MY_DIR}/${BACKEND_NAME}.py" ]; then
-        exec python ${MY_DIR}/${BACKEND_NAME}.py $@
+        python ${MY_DIR}/${BACKEND_NAME}.py $@
     fi
 }
 
@@ -210,4 +210,4 @@ function checkTargets() {
     echo false
 }
 
-init
+init

backend/python/common/template/requirements.txt

@@ -1,2 +1,2 @@
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf

backend/python/coqui/requirements-intel.txt

@@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/coqui/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 TTS==0.22.0
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 certifi
 transformers

backend/python/diffusers/backend.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 from concurrent import futures
-import traceback
+
 import argparse
 from collections import defaultdict
 from enum import Enum
@@ -17,39 +17,35 @@ import backend_pb2_grpc
 
 import grpc
 
-from diffusers import StableDiffusion3Pipeline, StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, \
-    EulerAncestralDiscreteScheduler
+from diffusers import StableDiffusion3Pipeline, StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, EulerAncestralDiscreteScheduler
 from diffusers import StableDiffusionImg2ImgPipeline, AutoPipelineForText2Image, ControlNetModel, StableVideoDiffusionPipeline
 from diffusers.pipelines.stable_diffusion import safety_checker
-from diffusers.utils import load_image, export_to_video
+from diffusers.utils import load_image,export_to_video
 from compel import Compel, ReturnedEmbeddingsType
 
 from transformers import CLIPTextModel
 from safetensors.torch import load_file
 
+
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
-COMPEL = os.environ.get("COMPEL", "0") == "1"
-XPU = os.environ.get("XPU", "0") == "1"
-CLIPSKIP = os.environ.get("CLIPSKIP", "1") == "1"
-SAFETENSORS = os.environ.get("SAFETENSORS", "1") == "1"
-CHUNK_SIZE = os.environ.get("CHUNK_SIZE", "8")
-FPS = os.environ.get("FPS", "7")
-DISABLE_CPU_OFFLOAD = os.environ.get("DISABLE_CPU_OFFLOAD", "0") == "1"
-FRAMES = os.environ.get("FRAMES", "64")
+COMPEL=os.environ.get("COMPEL", "0") == "1"
+XPU=os.environ.get("XPU", "0") == "1"
+CLIPSKIP=os.environ.get("CLIPSKIP", "1") == "1"
+SAFETENSORS=os.environ.get("SAFETENSORS", "1") == "1"
+CHUNK_SIZE=os.environ.get("CHUNK_SIZE", "8")
+FPS=os.environ.get("FPS", "7")
+DISABLE_CPU_OFFLOAD=os.environ.get("DISABLE_CPU_OFFLOAD", "0") == "1"
+FRAMES=os.environ.get("FRAMES", "64")
 
 if XPU:
     import intel_extension_for_pytorch as ipex
-
     print(ipex.xpu.get_device_name(0))
 
 # If MAX_WORKERS are specified in the environment use it, otherwise default to 1
 MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
 
-
 # https://github.com/CompVis/stable-diffusion/issues/239#issuecomment-1627615287
-def sc(self, clip_input, images): return images, [False for i in images]
-
-
+def sc(self, clip_input, images) : return images, [False for i in images]
 # edit the StableDiffusionSafetyChecker class so that, when called, it just returns the images and an array of True values
 safety_checker.StableDiffusionSafetyChecker.forward = sc
 
@@ -66,8 +62,6 @@ from diffusers.schedulers import (
     PNDMScheduler,
     UniPCMultistepScheduler,
 )
-
-
 # The scheduler list mapping was taken from here: https://github.com/neggles/animatediff-cli/blob/6f336f5f4b5e38e85d7f06f1744ef42d0a45f2a7/src/animatediff/schedulers.py#L39
 # Credits to https://github.com/neggles
 # See https://github.com/huggingface/diffusers/issues/4167 for more details on sched mapping from A1111
@@ -142,12 +136,10 @@ def get_scheduler(name: str, config: dict = {}):
 
     return sched_class.from_config(config)
 
-
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
     def Health(self, request, context):
         return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
-
     def LoadModel(self, request, context):
         try:
             print(f"Loading model {request.Model}...", file=sys.stderr)
@@ -157,7 +149,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
             if request.F16Memory:
                 torchType = torch.float16
-                variant = "fp16"
+                variant="fp16"
 
             local = False
             modelFile = request.Model
@@ -165,38 +157,38 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             self.cfg_scale = 7
             if request.CFGScale != 0:
                 self.cfg_scale = request.CFGScale
-
+            
             clipmodel = "runwayml/stable-diffusion-v1-5"
             if request.CLIPModel != "":
                 clipmodel = request.CLIPModel
             clipsubfolder = "text_encoder"
             if request.CLIPSubfolder != "":
                 clipsubfolder = request.CLIPSubfolder
-
+            
             # Check if ModelFile exists
             if request.ModelFile != "":
                 if os.path.exists(request.ModelFile):
                     local = True
                     modelFile = request.ModelFile
-
+            
             fromSingleFile = request.Model.startswith("http") or request.Model.startswith("/") or local
-            self.img2vid = False
-            self.txt2vid = False
+            self.img2vid=False
+            self.txt2vid=False
             ## img2img
             if (request.PipelineType == "StableDiffusionImg2ImgPipeline") or (request.IMG2IMG and request.PipelineType == ""):
                 if fromSingleFile:
                     self.pipe = StableDiffusionImg2ImgPipeline.from_single_file(modelFile,
-                                                                                torch_dtype=torchType)
+                                torch_dtype=torchType)
                 else:
                     self.pipe = StableDiffusionImg2ImgPipeline.from_pretrained(request.Model,
-                                                                               torch_dtype=torchType)
+                                torch_dtype=torchType)
 
             elif request.PipelineType == "StableDiffusionDepth2ImgPipeline":
                 self.pipe = StableDiffusionDepth2ImgPipeline.from_pretrained(request.Model,
-                                                                             torch_dtype=torchType)
+                            torch_dtype=torchType)
             ## img2vid
             elif request.PipelineType == "StableVideoDiffusionPipeline":
-                self.img2vid = True
+                self.img2vid=True
                 self.pipe = StableVideoDiffusionPipeline.from_pretrained(
                     request.Model, torch_dtype=torchType, variant=variant
                 )
@@ -205,63 +197,64 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             ## text2img
             elif request.PipelineType == "AutoPipelineForText2Image" or request.PipelineType == "":
                 self.pipe = AutoPipelineForText2Image.from_pretrained(request.Model,
-                                                                      torch_dtype=torchType,
-                                                                      use_safetensors=SAFETENSORS,
-                                                                      variant=variant)
+                                                    torch_dtype=torchType,
+                                                    use_safetensors=SAFETENSORS, 
+                                                    variant=variant)
             elif request.PipelineType == "StableDiffusionPipeline":
                 if fromSingleFile:
                     self.pipe = StableDiffusionPipeline.from_single_file(modelFile,
-                                                                         torch_dtype=torchType)
+                                                        torch_dtype=torchType)
                 else:
                     self.pipe = StableDiffusionPipeline.from_pretrained(request.Model,
-                                                                        torch_dtype=torchType)
+                                                        torch_dtype=torchType)
             elif request.PipelineType == "DiffusionPipeline":
                 self.pipe = DiffusionPipeline.from_pretrained(request.Model,
-                                                              torch_dtype=torchType)
+                                                        torch_dtype=torchType)
             elif request.PipelineType == "VideoDiffusionPipeline":
-                self.txt2vid = True
+                self.txt2vid=True
                 self.pipe = DiffusionPipeline.from_pretrained(request.Model,
-                                                              torch_dtype=torchType)
+                                                        torch_dtype=torchType)
             elif request.PipelineType == "StableDiffusionXLPipeline":
                 if fromSingleFile:
                     self.pipe = StableDiffusionXLPipeline.from_single_file(modelFile,
-                                                                           torch_dtype=torchType,
-                                                                           use_safetensors=True)
+                                                               torch_dtype=torchType,
+                                                               use_safetensors=True)
                 else:
                     self.pipe = StableDiffusionXLPipeline.from_pretrained(
-                        request.Model,
-                        torch_dtype=torchType,
-                        use_safetensors=True,
+                        request.Model, 
+                        torch_dtype=torchType, 
+                        use_safetensors=True, 
                         variant=variant)
             elif request.PipelineType == "StableDiffusion3Pipeline":
                 if fromSingleFile:
                     self.pipe = StableDiffusion3Pipeline.from_single_file(modelFile,
-                                                                          torch_dtype=torchType,
-                                                                          use_safetensors=True)
+                                                               torch_dtype=torchType,
+                                                               use_safetensors=True)
                 else:
                     self.pipe = StableDiffusion3Pipeline.from_pretrained(
-                        request.Model,
-                        torch_dtype=torchType,
-                        use_safetensors=True,
+                        request.Model, 
+                        torch_dtype=torchType, 
+                        use_safetensors=True, 
                         variant=variant)
 
             if CLIPSKIP and request.CLIPSkip != 0:
                 self.clip_skip = request.CLIPSkip
             else:
                 self.clip_skip = 0
-
+            
             # torch_dtype needs to be customized. float16 for GPU, float32 for CPU
             # TODO: this needs to be customized
             if request.SchedulerType != "":
                 self.pipe.scheduler = get_scheduler(request.SchedulerType, self.pipe.scheduler.config)
-
+                
             if COMPEL:
                 self.compel = Compel(
-                    tokenizer=[self.pipe.tokenizer, self.pipe.tokenizer_2],
+                    tokenizer=[self.pipe.tokenizer, self.pipe.tokenizer_2 ], 
                     text_encoder=[self.pipe.text_encoder, self.pipe.text_encoder_2],
                     returned_embeddings_type=ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NON_NORMALIZED,
                     requires_pooled=[False, True]
-                )
+                    )
+
 
             if request.ControlNet:
                 self.controlnet = ControlNetModel.from_pretrained(
@@ -270,6 +263,13 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                 self.pipe.controlnet = self.controlnet
             else:
                 self.controlnet = None
+
+            if request.CUDA:
+                self.pipe.to('cuda')
+                if self.controlnet:
+                    self.controlnet.to('cuda')
+            if XPU:
+                self.pipe = self.pipe.to("xpu")
             # Assume directory from request.ModelFile.
             # Only if request.LoraAdapter it's not an absolute path
             if request.LoraAdapter and request.ModelFile != "" and not os.path.isabs(request.LoraAdapter) and request.LoraAdapter:
@@ -282,17 +282,10 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             if request.LoraAdapter:
                 # Check if its a local file and not a directory ( we load lora differently for a safetensor file )
                 if os.path.exists(request.LoraAdapter) and not os.path.isdir(request.LoraAdapter):
-                    # self.load_lora_weights(request.LoraAdapter, 1, device, torchType)
-                    self.pipe.load_lora_weights(request.LoraAdapter)
+                    self.load_lora_weights(request.LoraAdapter, 1, device, torchType)
                 else:
                     self.pipe.unet.load_attn_procs(request.LoraAdapter)
 
-            if request.CUDA:
-                self.pipe.to('cuda')
-                if self.controlnet:
-                    self.controlnet.to('cuda')
-            if XPU:
-                self.pipe = self.pipe.to("xpu")
         except Exception as err:
             return backend_pb2.Result(success=False, message=f"Unexpected {err=}, {type(err)=}")
         # Implement your logic here for the LoadModel service
@@ -365,9 +358,9 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
         # create a dictionary of values for the parameters
         options = {
-            "negative_prompt": request.negative_prompt,
-            "width": request.width,
-            "height": request.height,
+            "negative_prompt":     request.negative_prompt, 
+            "width":               request.width, 
+            "height":              request.height,
             "num_inference_steps": steps,
         }
 
@@ -379,7 +372,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             options["image"] = pose_image
 
         if CLIPSKIP and self.clip_skip != 0:
-            options["clip_skip"] = self.clip_skip
+            options["clip_skip"]=self.clip_skip
 
         # Get the keys that we will build the args for our pipe for
         keys = options.keys()
@@ -423,21 +416,20 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             image = self.pipe(
                 guidance_scale=self.cfg_scale,
                 **kwargs
-            ).images[0]
+                ).images[0] 
         else:
             # pass the kwargs dictionary to the self.pipe method
             image = self.pipe(
                 prompt,
                 guidance_scale=self.cfg_scale,
                 **kwargs
-            ).images[0]
+                ).images[0]
 
         # save the result
         image.save(request.dst)
 
         return backend_pb2.Result(message="Media generated", success=True)
 
-
 def serve(address):
     server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
@@ -461,7 +453,6 @@ def serve(address):
     except KeyboardInterrupt:
         server.stop(0)
 
-
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(description="Run the gRPC server.")
     parser.add_argument(
@@ -469,4 +460,4 @@ if __name__ == "__main__":
     )
     args = parser.parse_args()
 
-    serve(args.addr)
+    serve(args.addr)

backend/python/diffusers/requirements-intel.txt

@@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchvision
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/diffusers/requirements.txt

@@ -1,9 +1,7 @@
-setuptools
 accelerate
 compel
-peft
 diffusers
-grpcio==1.65.0
+grpcio==1.64.0
 opencv-python
 pillow
 protobuf

backend/python/exllama/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 torch
 transformers

backend/python/exllama2/requirements.txt

@@ -1,5 +1,5 @@
 accelerate
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 certifi
 torch

backend/python/mamba/requirements-install.txt

@@ -4,4 +4,4 @@
 packaging
 setuptools
 wheel
-torch==2.3.1
+torch==2.2.0

backend/python/mamba/requirements.txt

@@ -1,6 +1,6 @@
-causal-conv1d==1.4.0
-mamba-ssm==2.2.2
-grpcio==1.65.0
+causal-conv1d==1.2.0.post2
+mamba-ssm==1.2.0.post1
+grpcio==1.64.0
 protobuf
 certifi
 transformers

backend/python/openvoice/requirements-intel.txt

@@ -2,22 +2,22 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-grpcio==1.64.1
+grpcio==1.64.0
 protobuf
 librosa==0.9.1
-faster-whisper==1.0.3
+faster-whisper==0.9.0
 pydub==0.25.1
 wavmark==0.0.3
-numpy==1.26.4
+numpy==1.22.0
 eng_to_ipa==0.0.2
 inflect==7.0.0
 unidecode==1.3.7
-whisper-timestamped==1.15.4
+whisper-timestamped==1.14.2
 openai
 python-dotenv
 pypinyin==0.50.0
 cn2an==0.5.22
 jieba==0.42.1
-gradio==4.38.1
+gradio==3.48.0
 langid==1.1.6
-git+https://github.com/myshell-ai/MeloTTS.git
+git+https://github.com/myshell-ai/MeloTTS.git

backend/python/openvoice/requirements.txt

@@ -1,20 +1,20 @@
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
-librosa
-faster-whisper
+librosa==0.9.1
+faster-whisper==0.9.0
 pydub==0.25.1
 wavmark==0.0.3
-numpy
+numpy==1.22.0
 eng_to_ipa==0.0.2
-inflect
-unidecode
-whisper-timestamped
+inflect==7.0.0
+unidecode==1.3.7
+whisper-timestamped==1.14.2
 openai
 python-dotenv
-pypinyin
+pypinyin==0.50.0
 cn2an==0.5.22
 jieba==0.42.1
-gradio
+gradio==3.48.0
 langid==1.1.6
 git+https://github.com/myshell-ai/MeloTTS.git
-git+https://github.com/myshell-ai/OpenVoice.git
+git+https://github.com/myshell-ai/OpenVoice.git

backend/python/parler-tts/requirements-intel.txt

@@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/parler-tts/requirements.txt

@@ -1,5 +1,5 @@
 accelerate
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 torch
 git+https://github.com/huggingface/parler-tts.git@10016fb0300c0dc31a0fb70e26f3affee7b62f16

backend/python/petals/requirements-intel.txt

@@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/rerankers/requirements-intel.txt

@@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/rerankers/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 rerankers[transformers]
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 certifi
 transformers

backend/python/sentencetransformers/requirements-intel.txt

@@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/sentencetransformers/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
-sentence-transformers==3.0.1
+sentence-transformers==2.5.1
 transformers
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 certifi

backend/python/transformers-musicgen/requirements-intel.txt

@@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/transformers-musicgen/requirements.txt

@@ -1,7 +1,7 @@
 accelerate
 transformers
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 torch
-scipy==1.14.0
+scipy==1.13.0
 certifi

backend/python/transformers/requirements-intel.txt

@@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/transformers/requirements.txt

@@ -1,9 +1,9 @@
 accelerate
 transformers
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 torch
 certifi
 intel-extension-for-transformers
 bitsandbytes
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/vall-e-x/requirements-intel.txt

@@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/vall-e-x/requirements.txt

@@ -1,4 +1,4 @@
 accelerate
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 certifi

backend/python/vllm/requirements-intel.txt

@@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/vllm/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 vllm
-grpcio==1.65.0
+grpcio==1.64.0
 protobuf
 certifi
 transformers

core/application.go

@@ -28,6 +28,7 @@ type Application struct {
 	// LocalAI System Services
 	BackendMonitorService *services.BackendMonitorService
 	GalleryService        *services.GalleryService
+	ListModelsService     *services.ListModelsService
 	LocalAIMetricsService *services.LocalAIMetricsService
 	// OpenAIService         *services.OpenAIService
 }

core/backend/llm.go

@@ -57,7 +57,7 @@ func ModelInference(ctx context.Context, s string, messages []schema.Message, im
 		if _, err := os.Stat(modelFile); os.IsNotExist(err) {
 			utils.ResetDownloadTimers()
 			// if we failed to load the model, we try to download it
-			err := gallery.InstallModelFromGallery(o.Galleries, modelFile, loader.ModelPath, gallery.GalleryModel{}, utils.DisplayDownloadFunction, o.EnforcePredownloadScans)
+			err := gallery.InstallModelFromGallery(o.Galleries, modelFile, loader.ModelPath, gallery.GalleryModel{}, utils.DisplayDownloadFunction)
 			if err != nil {
 				return nil, err
 			}

core/backend/options.go

@@ -91,7 +91,7 @@ func gRPCModelOpts(c config.BackendConfig) *pb.ModelOptions {
 		Type:                 c.ModelType,
 		RopeFreqScale:        c.RopeFreqScale,
 		NUMA:                 c.NUMA,
-		Embeddings:           *c.Embeddings,
+		Embeddings:           c.Embeddings,
 		LowVRAM:              *c.LowVRAM,
 		NGPULayers:           int32(*c.NGPULayers),
 		MMap:                 *c.MMap,
@@ -142,14 +142,12 @@ func gRPCPredictOpts(c config.BackendConfig, modelPath string) *pb.PredictOption
 		MirostatTAU:         float32(*c.LLMConfig.MirostatTAU),
 		Debug:               *c.Debug,
 		StopPrompts:         c.StopWords,
-		Repeat:              int32(c.RepeatLastN),
-		FrequencyPenalty:    float32(c.FrequencyPenalty),
-		PresencePenalty:     float32(c.PresencePenalty),
-		Penalty:             float32(c.RepeatPenalty),
+		Repeat:              int32(c.RepeatPenalty),
 		NKeep:               int32(c.Keep),
 		Batch:               int32(c.Batch),
 		IgnoreEOS:           c.IgnoreEOS,
 		Seed:                getSeed(c),
+		FrequencyPenalty:    float32(c.FrequencyPenalty),
 		MLock:               *c.MMlock,
 		MMap:                *c.MMap,
 		MainGPU:             c.MainGPU,

core/cli/cli.go

@@ -9,7 +9,6 @@ var CLI struct {
 	cliContext.Context `embed:""`
 
 	Run        RunCMD        `cmd:"" help:"Run LocalAI, this the default command if no other command is specified. Run 'local-ai run --help' for more information" default:"withargs"`
-	Federated  FederatedCLI  `cmd:"" help:"Run LocalAI in federated mode"`
 	Models     ModelsCMD     `cmd:"" help:"Manage LocalAI models and definitions"`
 	TTS        TTSCMD        `cmd:"" help:"Convert text to speech"`
 	Transcript TranscriptCMD `cmd:"" help:"Convert audio to text"`

core/cli/federated.go

@@ -1,21 +0,0 @@
-package cli
-
-import (
-	"context"
-
-	cliContext "github.com/mudler/LocalAI/core/cli/context"
-	"github.com/mudler/LocalAI/core/p2p"
-)
-
-type FederatedCLI struct {
-	Address        string `env:"LOCALAI_ADDRESS,ADDRESS" default:":8080" help:"Bind address for the API server" group:"api"`
-	Peer2PeerToken string `env:"LOCALAI_P2P_TOKEN,P2P_TOKEN,TOKEN" name:"p2ptoken" help:"Token for P2P mode (optional)" group:"p2p"`
-	LoadBalanced   bool   `env:"LOCALAI_LOAD_BALANCED,LOAD_BALANCED" default:"false" help:"Enable load balancing" group:"p2p"`
-}
-
-func (f *FederatedCLI) Run(ctx *cliContext.Context) error {
-
-	fs := p2p.NewFederatedServer(f.Address, p2p.FederatedID, f.Peer2PeerToken, f.LoadBalanced)
-
-	return fs.Start(context.Background())
-}

core/cli/models.go

@@ -2,7 +2,6 @@ package cli
 
 import (
 	"encoding/json"
-	"errors"
 	"fmt"
 
 	cliContext "github.com/mudler/LocalAI/core/cli/context"
@@ -25,8 +24,7 @@ type ModelsList struct {
 }
 
 type ModelsInstall struct {
-	DisablePredownloadScan bool     `env:"LOCALAI_DISABLE_PREDOWNLOAD_SCAN" help:"If true, disables the best-effort security scanner before downloading any files." group:"hardening" default:"false"`
-	ModelArgs              []string `arg:"" optional:"" name:"models" help:"Model configuration URLs to load"`
+	ModelArgs []string `arg:"" optional:"" name:"models" help:"Model configuration URLs to load"`
 
 	ModelsCMDFlags `embed:""`
 }
@@ -90,15 +88,9 @@ func (mi *ModelsInstall) Run(ctx *cliContext.Context) error {
 				return err
 			}
 
-			err = gallery.SafetyScanGalleryModel(model)
-			if err != nil && !errors.Is(err, downloader.ErrNonHuggingFaceFile) {
-				return err
-			}
-
 			log.Info().Str("model", modelName).Str("license", model.License).Msg("installing model")
 		}
-
-		err = startup.InstallModels(galleries, "", mi.ModelsPath, !mi.DisablePredownloadScan, progressCallback, modelName)
+		err = startup.InstallModels(galleries, "", mi.ModelsPath, progressCallback, modelName)
 		if err != nil {
 			return err
 		}

core/cli/run.go

@@ -3,8 +3,6 @@ package cli
 import (
 	"context"
 	"fmt"
-	"net"
-	"os"
 	"strings"
 	"time"
 
@@ -42,27 +40,25 @@ type RunCMD struct {
 	Threads     int  `env:"LOCALAI_THREADS,THREADS" short:"t" help:"Number of threads used for parallel computation. Usage of the number of physical cores in the system is suggested" group:"performance"`
 	ContextSize int  `env:"LOCALAI_CONTEXT_SIZE,CONTEXT_SIZE" default:"512" help:"Default context size for models" group:"performance"`
 
-	Address                string   `env:"LOCALAI_ADDRESS,ADDRESS" default:":8080" help:"Bind address for the API server" group:"api"`
-	CORS                   bool     `env:"LOCALAI_CORS,CORS" help:"" group:"api"`
-	CORSAllowOrigins       string   `env:"LOCALAI_CORS_ALLOW_ORIGINS,CORS_ALLOW_ORIGINS" group:"api"`
-	LibraryPath            string   `env:"LOCALAI_LIBRARY_PATH,LIBRARY_PATH" help:"Path to the library directory (for e.g. external libraries used by backends)" default:"/usr/share/local-ai/libs" group:"backends"`
-	CSRF                   bool     `env:"LOCALAI_CSRF" help:"Enables fiber CSRF middleware" group:"api"`
-	UploadLimit            int      `env:"LOCALAI_UPLOAD_LIMIT,UPLOAD_LIMIT" default:"15" help:"Default upload-limit in MB" group:"api"`
-	APIKeys                []string `env:"LOCALAI_API_KEY,API_KEY" help:"List of API Keys to enable API authentication. When this is set, all the requests must be authenticated with one of these API keys" group:"api"`
-	DisableWebUI           bool     `env:"LOCALAI_DISABLE_WEBUI,DISABLE_WEBUI" default:"false" help:"Disable webui" group:"api"`
-	DisablePredownloadScan bool     `env:"LOCALAI_DISABLE_PREDOWNLOAD_SCAN" help:"If true, disables the best-effort security scanner before downloading any files." group:"hardening" default:"false"`
-	OpaqueErrors           bool     `env:"LOCALAI_OPAQUE_ERRORS" default:"false" help:"If true, all error responses are replaced with blank 500 errors. This is intended only for hardening against information leaks and is normally not recommended." group:"hardening"`
-	Peer2Peer              bool     `env:"LOCALAI_P2P,P2P" name:"p2p" default:"false" help:"Enable P2P mode" group:"p2p"`
-	Peer2PeerToken         string   `env:"LOCALAI_P2P_TOKEN,P2P_TOKEN,TOKEN" name:"p2ptoken" help:"Token for P2P mode (optional)" group:"p2p"`
-	ParallelRequests       bool     `env:"LOCALAI_PARALLEL_REQUESTS,PARALLEL_REQUESTS" help:"Enable backends to handle multiple requests in parallel if they support it (e.g.: llama.cpp or vllm)" group:"backends"`
-	SingleActiveBackend    bool     `env:"LOCALAI_SINGLE_ACTIVE_BACKEND,SINGLE_ACTIVE_BACKEND" help:"Allow only one backend to be run at a time" group:"backends"`
-	PreloadBackendOnly     bool     `env:"LOCALAI_PRELOAD_BACKEND_ONLY,PRELOAD_BACKEND_ONLY" default:"false" help:"Do not launch the API services, only the preloaded models / backends are started (useful for multi-node setups)" group:"backends"`
-	ExternalGRPCBackends   []string `env:"LOCALAI_EXTERNAL_GRPC_BACKENDS,EXTERNAL_GRPC_BACKENDS" help:"A list of external grpc backends" group:"backends"`
-	EnableWatchdogIdle     bool     `env:"LOCALAI_WATCHDOG_IDLE,WATCHDOG_IDLE" default:"false" help:"Enable watchdog for stopping backends that are idle longer than the watchdog-idle-timeout" group:"backends"`
-	WatchdogIdleTimeout    string   `env:"LOCALAI_WATCHDOG_IDLE_TIMEOUT,WATCHDOG_IDLE_TIMEOUT" default:"15m" help:"Threshold beyond which an idle backend should be stopped" group:"backends"`
-	EnableWatchdogBusy     bool     `env:"LOCALAI_WATCHDOG_BUSY,WATCHDOG_BUSY" default:"false" help:"Enable watchdog for stopping backends that are busy longer than the watchdog-busy-timeout" group:"backends"`
-	WatchdogBusyTimeout    string   `env:"LOCALAI_WATCHDOG_BUSY_TIMEOUT,WATCHDOG_BUSY_TIMEOUT" default:"5m" help:"Threshold beyond which a busy backend should be stopped" group:"backends"`
-	Federated              bool     `env:"LOCALAI_FEDERATED,FEDERATED" help:"Enable federated instance" group:"federated"`
+	Address              string   `env:"LOCALAI_ADDRESS,ADDRESS" default:":8080" help:"Bind address for the API server" group:"api"`
+	CORS                 bool     `env:"LOCALAI_CORS,CORS" help:"" group:"api"`
+	CORSAllowOrigins     string   `env:"LOCALAI_CORS_ALLOW_ORIGINS,CORS_ALLOW_ORIGINS" group:"api"`
+	LibraryPath          string   `env:"LOCALAI_LIBRARY_PATH,LIBRARY_PATH" help:"Path to the library directory (for e.g. external libraries used by backends)" default:"/usr/share/local-ai/libs" group:"backends"`
+	CSRF                 bool     `env:"LOCALAI_CSRF" help:"Enables fiber CSRF middleware" group:"api"`
+	UploadLimit          int      `env:"LOCALAI_UPLOAD_LIMIT,UPLOAD_LIMIT" default:"15" help:"Default upload-limit in MB" group:"api"`
+	APIKeys              []string `env:"LOCALAI_API_KEY,API_KEY" help:"List of API Keys to enable API authentication. When this is set, all the requests must be authenticated with one of these API keys" group:"api"`
+	DisableWebUI         bool     `env:"LOCALAI_DISABLE_WEBUI,DISABLE_WEBUI" default:"false" help:"Disable webui" group:"api"`
+	OpaqueErrors         bool     `env:"LOCALAI_OPAQUE_ERRORS" default:"false" help:"If true, all error responses are replaced with blank 500 errors. This is intended only for hardening against information leaks and is normally not recommended." group:"api"`
+	Peer2Peer            bool     `env:"LOCALAI_P2P,P2P" name:"p2p" default:"false" help:"Enable P2P mode" group:"p2p"`
+	Peer2PeerToken       string   `env:"LOCALAI_P2P_TOKEN,P2P_TOKEN" name:"p2ptoken" help:"Token for P2P mode (optional)" group:"p2p"`
+	ParallelRequests     bool     `env:"LOCALAI_PARALLEL_REQUESTS,PARALLEL_REQUESTS" help:"Enable backends to handle multiple requests in parallel if they support it (e.g.: llama.cpp or vllm)" group:"backends"`
+	SingleActiveBackend  bool     `env:"LOCALAI_SINGLE_ACTIVE_BACKEND,SINGLE_ACTIVE_BACKEND" help:"Allow only one backend to be run at a time" group:"backends"`
+	PreloadBackendOnly   bool     `env:"LOCALAI_PRELOAD_BACKEND_ONLY,PRELOAD_BACKEND_ONLY" default:"false" help:"Do not launch the API services, only the preloaded models / backends are started (useful for multi-node setups)" group:"backends"`
+	ExternalGRPCBackends []string `env:"LOCALAI_EXTERNAL_GRPC_BACKENDS,EXTERNAL_GRPC_BACKENDS" help:"A list of external grpc backends" group:"backends"`
+	EnableWatchdogIdle   bool     `env:"LOCALAI_WATCHDOG_IDLE,WATCHDOG_IDLE" default:"false" help:"Enable watchdog for stopping backends that are idle longer than the watchdog-idle-timeout" group:"backends"`
+	WatchdogIdleTimeout  string   `env:"LOCALAI_WATCHDOG_IDLE_TIMEOUT,WATCHDOG_IDLE_TIMEOUT" default:"15m" help:"Threshold beyond which an idle backend should be stopped" group:"backends"`
+	EnableWatchdogBusy   bool     `env:"LOCALAI_WATCHDOG_BUSY,WATCHDOG_BUSY" default:"false" help:"Enable watchdog for stopping backends that are busy longer than the watchdog-busy-timeout" group:"backends"`
+	WatchdogBusyTimeout  string   `env:"LOCALAI_WATCHDOG_BUSY_TIMEOUT,WATCHDOG_BUSY_TIMEOUT" default:"5m" help:"Threshold beyond which a busy backend should be stopped" group:"backends"`
 }
 
 func (r *RunCMD) Run(ctx *cliContext.Context) error {
@@ -93,13 +89,11 @@ func (r *RunCMD) Run(ctx *cliContext.Context) error {
 		config.WithApiKeys(r.APIKeys),
 		config.WithModelsURL(append(r.Models, r.ModelArgs...)...),
 		config.WithOpaqueErrors(r.OpaqueErrors),
-		config.WithEnforcedPredownloadScans(!r.DisablePredownloadScan),
 	}
 
-	token := ""
 	if r.Peer2Peer || r.Peer2PeerToken != "" {
 		log.Info().Msg("P2P mode enabled")
-		token = r.Peer2PeerToken
+		token := r.Peer2PeerToken
 		if token == "" {
 			// IF no token is provided, and p2p is enabled,
 			// we generate one and wait for the user to pick up the token (this is for interactive)
@@ -110,46 +104,14 @@ func (r *RunCMD) Run(ctx *cliContext.Context) error {
 
 			log.Info().Msg("To use the token, you can run the following command in another node or terminal:")
 			fmt.Printf("export TOKEN=\"%s\"\nlocal-ai worker p2p-llama-cpp-rpc\n", token)
-		}
-		opts = append(opts, config.WithP2PToken(token))
 
-		node, err := p2p.NewNode(token)
-		if err != nil {
-			return err
+			// Ask for user confirmation
+			log.Info().Msg("Press a button to proceed")
+			var input string
+			fmt.Scanln(&input)
 		}
-
 		log.Info().Msg("Starting P2P server discovery...")
-		if err := p2p.ServiceDiscoverer(context.Background(), node, token, "", func(serviceID string, node p2p.NodeData) {
-			var tunnelAddresses []string
-			for _, v := range p2p.GetAvailableNodes("") {
-				if v.IsOnline() {
-					tunnelAddresses = append(tunnelAddresses, v.TunnelAddress)
-				} else {
-					log.Info().Msgf("Node %s is offline", v.ID)
-				}
-			}
-			tunnelEnvVar := strings.Join(tunnelAddresses, ",")
-
-			os.Setenv("LLAMACPP_GRPC_SERVERS", tunnelEnvVar)
-			log.Debug().Msgf("setting LLAMACPP_GRPC_SERVERS to %s", tunnelEnvVar)
-		}); err != nil {
-			return err
-		}
-	}
-
-	if r.Federated {
-		_, port, err := net.SplitHostPort(r.Address)
-		if err != nil {
-			return err
-		}
-		if err := p2p.ExposeService(context.Background(), "localhost", port, token, p2p.FederatedID); err != nil {
-			return err
-		}
-		node, err := p2p.NewNode(token)
-		if err != nil {
-			return err
-		}
-		if err := p2p.ServiceDiscoverer(context.Background(), node, token, p2p.FederatedID, nil); err != nil {
+		if err := p2p.LLamaCPPRPCServerDiscoverer(context.Background(), token); err != nil {
 			return err
 		}
 	}

core/cli/transcript.go

@@ -18,7 +18,7 @@ type TranscriptCMD struct {
 	Backend           string `short:"b" default:"whisper" help:"Backend to run the transcription model"`
 	Model             string `short:"m" required:"" help:"Model name to run the TTS"`
 	Language          string `short:"l" help:"Language of the audio file"`
-	Translate         bool   `short:"c" help:"Translate the transcription to english"`
+	Translate         bool   `short:"t" help:"Translate the transcription to english"`
 	Threads           int    `short:"t" default:"1" help:"Number of threads used for parallel computation"`
 	ModelsPath        string `env:"LOCALAI_MODELS_PATH,MODELS_PATH" type:"path" default:"${basepath}/models" help:"Path containing models used for inferencing" group:"storage"`
 	BackendAssetsPath string `env:"LOCALAI_BACKEND_ASSETS_PATH,BACKEND_ASSETS_PATH" type:"path" default:"/tmp/localai/backend_data" help:"Path used to extract libraries that are required by some of the backends in runtime" group:"storage"`

core/cli/util.go

@@ -1,22 +1,16 @@
 package cli
 
 import (
-	"encoding/json"
-	"errors"
 	"fmt"
 
 	"github.com/rs/zerolog/log"
 
 	cliContext "github.com/mudler/LocalAI/core/cli/context"
-	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/gallery"
-	"github.com/mudler/LocalAI/pkg/downloader"
 	gguf "github.com/thxcode/gguf-parser-go"
 )
 
 type UtilCMD struct {
 	GGUFInfo GGUFInfoCMD `cmd:"" name:"gguf-info" help:"Get information about a GGUF file"`
-	HFScan   HFScanCMD   `cmd:"" name:"hf-scan" help:"Checks installed models for known security issues. WARNING: this is a best-effort feature and may not catch everything!"`
 }
 
 type GGUFInfoCMD struct {
@@ -24,12 +18,6 @@ type GGUFInfoCMD struct {
 	Header bool     `optional:"" default:"false" name:"header" help:"Show header information"`
 }
 
-type HFScanCMD struct {
-	ModelsPath string   `env:"LOCALAI_MODELS_PATH,MODELS_PATH" type:"path" default:"${basepath}/models" help:"Path containing models used for inferencing" group:"storage"`
-	Galleries  string   `env:"LOCALAI_GALLERIES,GALLERIES" help:"JSON list of galleries" group:"models" default:"${galleries}"`
-	ToScan     []string `arg:""`
-}
-
 func (u *GGUFInfoCMD) Run(ctx *cliContext.Context) error {
 	if u.Args == nil || len(u.Args) == 0 {
 		return fmt.Errorf("no GGUF file provided")
@@ -65,37 +53,3 @@ func (u *GGUFInfoCMD) Run(ctx *cliContext.Context) error {
 
 	return nil
 }
-
-func (hfscmd *HFScanCMD) Run(ctx *cliContext.Context) error {
-	log.Info().Msg("LocalAI Security Scanner - This is BEST EFFORT functionality! Currently limited to huggingface models!")
-	if len(hfscmd.ToScan) == 0 {
-		log.Info().Msg("Checking all installed models against galleries")
-		var galleries []config.Gallery
-		if err := json.Unmarshal([]byte(hfscmd.Galleries), &galleries); err != nil {
-			log.Error().Err(err).Msg("unable to load galleries")
-		}
-
-		err := gallery.SafetyScanGalleryModels(galleries, hfscmd.ModelsPath)
-		if err == nil {
-			log.Info().Msg("No security warnings were detected for your installed models. Please note that this is a BEST EFFORT tool, and all issues may not be detected.")
-		} else {
-			log.Error().Err(err).Msg("! WARNING ! A known-vulnerable model is installed!")
-		}
-		return err
-	} else {
-		var errs error = nil
-		for _, uri := range hfscmd.ToScan {
-			log.Info().Str("uri", uri).Msg("scanning specific uri")
-			scanResults, err := downloader.HuggingFaceScan(uri)
-			if err != nil && !errors.Is(err, downloader.ErrNonHuggingFaceFile) {
-				log.Error().Err(err).Strs("clamAV", scanResults.ClamAVInfectedFiles).Strs("pickles", scanResults.DangerousPickles).Msg("! WARNING ! A known-vulnerable model is included in this repo!")
-				errs = errors.Join(errs, err)
-			}
-		}
-		if errs != nil {
-			return errs
-		}
-		log.Info().Msg("No security warnings were detected for your installed models. Please note that this is a BEST EFFORT tool, and all issues may not be detected.")
-		return nil
-	}
-}

core/cli/worker/worker_p2p.go

@@ -20,7 +20,7 @@ import (
 
 type P2P struct {
 	WorkerFlags       `embed:""`
-	Token             string   `env:"LOCALAI_TOKEN,LOCALAI_P2P_TOKEN,TOKEN" help:"P2P token to use"`
+	Token             string   `env:"LOCALAI_TOKEN,TOKEN" help:"JSON list of galleries"`
 	NoRunner          bool     `env:"LOCALAI_NO_RUNNER,NO_RUNNER" help:"Do not start the llama-cpp-rpc-server"`
 	RunnerAddress     string   `env:"LOCALAI_RUNNER_ADDRESS,RUNNER_ADDRESS" help:"Address of the llama-cpp-rpc-server"`
 	RunnerPort        string   `env:"LOCALAI_RUNNER_PORT,RUNNER_PORT" help:"Port of the llama-cpp-rpc-server"`
@@ -59,7 +59,7 @@ func (r *P2P) Run(ctx *cliContext.Context) error {
 			p = r.RunnerPort
 		}
 
-		err = p2p.ExposeService(context.Background(), address, p, r.Token, "")
+		err = p2p.BindLLamaCPPWorker(context.Background(), address, p, r.Token)
 		if err != nil {
 			return err
 		}
@@ -99,7 +99,7 @@ func (r *P2P) Run(ctx *cliContext.Context) error {
 		}
 	}()
 
-	err = p2p.ExposeService(context.Background(), address, fmt.Sprint(port), r.Token, "")
+	err = p2p.BindLLamaCPPWorker(context.Background(), address, fmt.Sprint(port), r.Token)
 	if err != nil {
 		return err
 	}

core/config/application_config.go

@@ -31,9 +31,7 @@ type ApplicationConfig struct {
 	PreloadModelsFromPath               string
 	CORSAllowOrigins                    string
 	ApiKeys                             []string
-	EnforcePredownloadScans             bool
 	OpaqueErrors                        bool
-	P2PToken                            string
 
 	ModelLibraryURL string
 
@@ -97,12 +95,6 @@ func WithCsrf(b bool) AppOption {
 	}
 }
 
-func WithP2PToken(s string) AppOption {
-	return func(o *ApplicationConfig) {
-		o.P2PToken = s
-	}
-}
-
 func WithModelLibraryURL(url string) AppOption {
 	return func(o *ApplicationConfig) {
 		o.ModelLibraryURL = url
@@ -302,12 +294,6 @@ func WithApiKeys(apiKeys []string) AppOption {
 	}
 }
 
-func WithEnforcedPredownloadScans(enforced bool) AppOption {
-	return func(o *ApplicationConfig) {
-		o.EnforcePredownloadScans = enforced
-	}
-}
-
 func WithOpaqueErrors(opaque bool) AppOption {
 	return func(o *ApplicationConfig) {
 		o.OpaqueErrors = opaque

core/config/backend_config.go

@@ -32,7 +32,7 @@ type BackendConfig struct {
 	Threads        *int              `yaml:"threads"`
 	Debug          *bool             `yaml:"debug"`
 	Roles          map[string]string `yaml:"roles"`
-	Embeddings     *bool             `yaml:"embeddings"`
+	Embeddings     bool              `yaml:"embeddings"`
 	Backend        string            `yaml:"backend"`
 	TemplateConfig TemplateConfig    `yaml:"template"`
 
@@ -338,10 +338,6 @@ func (cfg *BackendConfig) SetDefaults(opts ...ConfigLoaderOption) {
 		cfg.LowVRAM = &falseV
 	}
 
-	if cfg.Embeddings == nil {
-		cfg.Embeddings = &falseV
-	}
-
 	// Value passed by the top level are treated as default (no implicit defaults)
 	// defaults are set by the user
 	if ctx == 0 {

core/config/guesser.go

@@ -20,7 +20,6 @@ const (
 	ChatML
 	Mistral03
 	Gemma
-	DeepSeek2
 )
 
 type settingsConfig struct {
@@ -38,17 +37,6 @@ var defaultsSettings map[familyType]settingsConfig = map[familyType]settingsConf
 			Completion:  "{{.Input}}",
 		},
 	},
-	DeepSeek2: {
-		StopWords: []string{"<｜end▁of▁sentence｜>"},
-		TemplateConfig: TemplateConfig{
-			ChatMessage: `{{if eq .RoleName "user" -}}User: {{.Content }}
-{{ end -}}
-{{if eq .RoleName "assistant" -}}Assistant: {{.Content}}<｜end▁of▁sentence｜>{{end}}
-{{if eq .RoleName "system" -}}{{.Content}}
-{{end -}}`,
-			Chat: "{{.Input -}}\nAssistant: ",
-		},
-	},
 	LLaMa3: {
 		StopWords: []string{"<|eot_id|>"},
 		TemplateConfig: TemplateConfig{
@@ -220,11 +208,8 @@ func identifyFamily(f *gguf.GGUFFile) familyType {
 	qwen2 := arch == "qwen2"
 	phi3 := arch == "phi-3"
 	gemma := strings.HasPrefix(f.Model().Name, "gemma")
-	deepseek2 := arch == "deepseek2"
 
 	switch {
-	case deepseek2:
-		return DeepSeek2
 	case gemma:
 		return Gemma
 	case llama3:

core/gallery/gallery.go

@@ -7,16 +7,15 @@ import (
 	"path/filepath"
 	"strings"
 
-	"dario.cat/mergo"
+	"github.com/imdario/mergo"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/pkg/downloader"
-	"github.com/mudler/LocalAI/pkg/utils"
 	"github.com/rs/zerolog/log"
 	"gopkg.in/yaml.v2"
 )
 
 // Installs a model from the gallery
-func InstallModelFromGallery(galleries []config.Gallery, name string, basePath string, req GalleryModel, downloadStatus func(string, string, string, float64), enforceScan bool) error {
+func InstallModelFromGallery(galleries []config.Gallery, name string, basePath string, req GalleryModel, downloadStatus func(string, string, string, float64)) error {
 
 	applyModel := func(model *GalleryModel) error {
 		name = strings.ReplaceAll(name, string(os.PathSeparator), "__")
@@ -64,7 +63,7 @@ func InstallModelFromGallery(galleries []config.Gallery, name string, basePath s
 			return err
 		}
 
-		if err := InstallModel(basePath, installName, &config, model.Overrides, downloadStatus, enforceScan); err != nil {
+		if err := InstallModel(basePath, installName, &config, model.Overrides, downloadStatus); err != nil {
 			return err
 		}
 
@@ -190,12 +189,6 @@ func DeleteModelFromSystem(basePath string, name string, additionalFiles []strin
 
 	galleryFile := filepath.Join(basePath, galleryFileName(name))
 
-	for _, f := range []string{configFile, galleryFile} {
-		if err := utils.VerifyPath(f, basePath); err != nil {
-			return fmt.Errorf("failed to verify path %s: %w", f, err)
-		}
-	}
-
 	var err error
 	// Delete all the files associated to the model
 	// read the model config
@@ -235,29 +228,3 @@ func DeleteModelFromSystem(basePath string, name string, additionalFiles []strin
 
 	return err
 }
-
-// This is ***NEVER*** going to be perfect or finished.
-// This is a BEST EFFORT function to surface known-vulnerable models to users.
-func SafetyScanGalleryModels(galleries []config.Gallery, basePath string) error {
-	galleryModels, err := AvailableGalleryModels(galleries, basePath)
-	if err != nil {
-		return err
-	}
-	for _, gM := range galleryModels {
-		if gM.Installed {
-			err = errors.Join(err, SafetyScanGalleryModel(gM))
-		}
-	}
-	return err
-}
-
-func SafetyScanGalleryModel(galleryModel *GalleryModel) error {
-	for _, file := range galleryModel.AdditionalFiles {
-		scanResults, err := downloader.HuggingFaceScan(file.URI)
-		if err != nil && !errors.Is(err, downloader.ErrNonHuggingFaceFile) {
-			log.Error().Str("model", galleryModel.Name).Strs("clamAV", scanResults.ClamAVInfectedFiles).Strs("pickles", scanResults.DangerousPickles).Msg("Contains unsafe file(s)!")
-			return err
-		}
-	}
-	return nil
-}

core/gallery/models.go

@@ -1,12 +1,11 @@
 package gallery
 
 import (
-	"errors"
 	"fmt"
 	"os"
 	"path/filepath"
 
-	"dario.cat/mergo"
+	"github.com/imdario/mergo"
 	lconfig "github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/pkg/downloader"
 	"github.com/mudler/LocalAI/pkg/utils"
@@ -95,7 +94,7 @@ func ReadConfigFile(filePath string) (*Config, error) {
 	return &config, nil
 }
 
-func InstallModel(basePath, nameOverride string, config *Config, configOverrides map[string]interface{}, downloadStatus func(string, string, string, float64), enforceScan bool) error {
+func InstallModel(basePath, nameOverride string, config *Config, configOverrides map[string]interface{}, downloadStatus func(string, string, string, float64)) error {
 	// Create base path if it doesn't exist
 	err := os.MkdirAll(basePath, 0750)
 	if err != nil {
@@ -113,18 +112,9 @@ func InstallModel(basePath, nameOverride string, config *Config, configOverrides
 		if err := utils.VerifyPath(file.Filename, basePath); err != nil {
 			return err
 		}
-
 		// Create file path
 		filePath := filepath.Join(basePath, file.Filename)
 
-		if enforceScan {
-			scanResults, err := downloader.HuggingFaceScan(file.URI)
-			if err != nil && !errors.Is(err, downloader.ErrNonHuggingFaceFile) {
-				log.Error().Str("model", config.Name).Strs("clamAV", scanResults.ClamAVInfectedFiles).Strs("pickles", scanResults.DangerousPickles).Msg("Contains unsafe file(s)!")
-				return err
-			}
-		}
-
 		if err := downloader.DownloadFile(file.URI, filePath, file.SHA256, i, len(config.Files), downloadStatus); err != nil {
 			return err
 		}

core/gallery/models_test.go

@@ -21,7 +21,7 @@ var _ = Describe("Model test", func() {
 			defer os.RemoveAll(tempdir)
 			c, err := ReadConfigFile(filepath.Join(os.Getenv("FIXTURES"), "gallery_simple.yaml"))
 			Expect(err).ToNot(HaveOccurred())
-			err = InstallModel(tempdir, "", c, map[string]interface{}{}, func(string, string, string, float64) {}, true)
+			err = InstallModel(tempdir, "", c, map[string]interface{}{}, func(string, string, string, float64) {})
 			Expect(err).ToNot(HaveOccurred())
 
 			for _, f := range []string{"cerebras", "cerebras-completion.tmpl", "cerebras-chat.tmpl", "cerebras.yaml"} {
@@ -69,7 +69,7 @@ var _ = Describe("Model test", func() {
 			Expect(models[0].URL).To(Equal("https://raw.githubusercontent.com/go-skynet/model-gallery/main/bert-embeddings.yaml"))
 			Expect(models[0].Installed).To(BeFalse())
 
-			err = InstallModelFromGallery(galleries, "test@bert", tempdir, GalleryModel{}, func(s1, s2, s3 string, f float64) {}, true)
+			err = InstallModelFromGallery(galleries, "test@bert", tempdir, GalleryModel{}, func(s1, s2, s3 string, f float64) {})
 			Expect(err).ToNot(HaveOccurred())
 
 			dat, err := os.ReadFile(filepath.Join(tempdir, "bert.yaml"))
@@ -106,7 +106,7 @@ var _ = Describe("Model test", func() {
 			c, err := ReadConfigFile(filepath.Join(os.Getenv("FIXTURES"), "gallery_simple.yaml"))
 			Expect(err).ToNot(HaveOccurred())
 
-			err = InstallModel(tempdir, "foo", c, map[string]interface{}{}, func(string, string, string, float64) {}, true)
+			err = InstallModel(tempdir, "foo", c, map[string]interface{}{}, func(string, string, string, float64) {})
 			Expect(err).ToNot(HaveOccurred())
 
 			for _, f := range []string{"cerebras", "cerebras-completion.tmpl", "cerebras-chat.tmpl", "foo.yaml"} {
@@ -122,7 +122,7 @@ var _ = Describe("Model test", func() {
 			c, err := ReadConfigFile(filepath.Join(os.Getenv("FIXTURES"), "gallery_simple.yaml"))
 			Expect(err).ToNot(HaveOccurred())
 
-			err = InstallModel(tempdir, "foo", c, map[string]interface{}{"backend": "foo"}, func(string, string, string, float64) {}, true)
+			err = InstallModel(tempdir, "foo", c, map[string]interface{}{"backend": "foo"}, func(string, string, string, float64) {})
 			Expect(err).ToNot(HaveOccurred())
 
 			for _, f := range []string{"cerebras", "cerebras-completion.tmpl", "cerebras-chat.tmpl", "foo.yaml"} {
@@ -148,7 +148,7 @@ var _ = Describe("Model test", func() {
 			c, err := ReadConfigFile(filepath.Join(os.Getenv("FIXTURES"), "gallery_simple.yaml"))
 			Expect(err).ToNot(HaveOccurred())
 
-			err = InstallModel(tempdir, "../../../foo", c, map[string]interface{}{}, func(string, string, string, float64) {}, true)
+			err = InstallModel(tempdir, "../../../foo", c, map[string]interface{}{}, func(string, string, string, float64) {})
 			Expect(err).To(HaveOccurred())
 		})
 	})

core/http/ctx/fiber.go

@@ -5,8 +5,6 @@ import (
 	"strings"
 
 	"github.com/gofiber/fiber/v2"
-	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
@@ -15,7 +13,7 @@ import (
 // If no model is specified, it will take the first available
 // Takes a model string as input which should be the one received from the user request.
 // It returns the model name resolved from the context and an error if any.
-func ModelFromContext(ctx *fiber.Ctx, cl *config.BackendConfigLoader, loader *model.ModelLoader, modelInput string, firstModel bool) (string, error) {
+func ModelFromContext(ctx *fiber.Ctx, loader *model.ModelLoader, modelInput string, firstModel bool) (string, error) {
 	if ctx.Params("model") != "" {
 		modelInput = ctx.Params("model")
 	}
@@ -26,7 +24,7 @@ func ModelFromContext(ctx *fiber.Ctx, cl *config.BackendConfigLoader, loader *mo
 
 	// If no model was specified, take the first available
 	if modelInput == "" && !bearerExists && firstModel {
-		models, _ := services.ListModels(cl, loader, "", true)
+		models, _ := loader.ListModels()
 		if len(models) > 0 {
 			modelInput = models[0]
 			log.Debug().Msgf("No model specified, using: %s", modelInput)

core/http/elements/gallery.go

@@ -7,7 +7,6 @@ import (
 	"github.com/chasefleming/elem-go"
 	"github.com/chasefleming/elem-go/attrs"
 	"github.com/mudler/LocalAI/core/gallery"
-	"github.com/mudler/LocalAI/core/p2p"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/pkg/xsync"
 )
@@ -16,14 +15,6 @@ const (
 	noImage = "https://upload.wikimedia.org/wikipedia/commons/6/65/No-Image-Placeholder.svg"
 )
 
-func renderElements(n []elem.Node) string {
-	render := ""
-	for _, r := range n {
-		render += r.Render()
-	}
-	return render
-}
-
 func DoneProgress(galleryID, text string, showDelete bool) string {
 	var modelName = galleryID
 	// Split by @ and grab the name
@@ -81,135 +72,6 @@ func ProgressBar(progress string) string {
 	).Render()
 }
 
-func P2PNodeStats(nodes []p2p.NodeData) string {
-	/*
-	   <div class="bg-gray-800 p-6 rounded-lg shadow-lg text-left">
-	                       <p class="text-xl font-semibold text-gray-200">Total Workers Detected: {{ len .Nodes }}</p>
-	                       {{ $online := 0 }}
-	                       {{ range .Nodes }}
-	                           {{ if .IsOnline }}
-	                               {{ $online = add $online 1 }}
-	                           {{ end }}
-	                       {{ end }}
-	                       <p class="text-xl font-semibold text-gray-200">Total Online Workers: {{$online}}</p>
-	                   </div>
-	*/
-
-	online := 0
-	for _, n := range nodes {
-		if n.IsOnline() {
-			online++
-		}
-	}
-
-	class := "text-green-500"
-	if online == 0 {
-		class = "text-red-500"
-	}
-	/*
-	   <i class="fas fa-circle animate-pulse text-green-500 ml-2 mr-1"></i>
-	*/
-	circle := elem.I(attrs.Props{
-		"class": "fas fa-circle animate-pulse " + class + " ml-2 mr-1",
-	})
-	nodesElements := []elem.Node{
-		elem.Span(
-			attrs.Props{
-				"class": class,
-			},
-			circle,
-			elem.Text(fmt.Sprintf("%d", online)),
-		),
-		elem.Span(
-			attrs.Props{
-				"class": "text-gray-200",
-			},
-			elem.Text(fmt.Sprintf("/%d", len(nodes))),
-		),
-	}
-
-	return renderElements(nodesElements)
-}
-
-func P2PNodeBoxes(nodes []p2p.NodeData) string {
-	/*
-			<div class="bg-gray-800 p-4 rounded-lg shadow-lg text-left">
-			<div class="flex items-center mb-2">
-				<i class="fas fa-desktop text-gray-400 mr-2"></i>
-				<span class="text-gray-200 font-semibold">{{.ID}}</span>
-			</div>
-			<p class="text-sm text-gray-400 mt-2 flex items-center">
-				Status:
-				<i class="fas fa-circle {{ if .IsOnline }}text-green-500{{ else }}text-red-500{{ end }} ml-2 mr-1"></i>
-				<span class="{{ if .IsOnline }}text-green-400{{ else }}text-red-400{{ end }}">
-					{{ if .IsOnline }}Online{{ else }}Offline{{ end }}
-				</span>
-			</p>
-		</div>
-	*/
-
-	nodesElements := []elem.Node{}
-
-	for _, n := range nodes {
-
-		nodesElements = append(nodesElements,
-			elem.Div(
-				attrs.Props{
-					"class": "bg-gray-700 p-6 rounded-lg shadow-lg text-left",
-				},
-				elem.P(
-					attrs.Props{
-						"class": "text-sm text-gray-400 mt-2 flex",
-					},
-					elem.I(
-						attrs.Props{
-							"class": "fas fa-desktop text-gray-400 mr-2",
-						},
-					),
-					elem.Text("Name: "),
-					elem.Span(
-						attrs.Props{
-							"class": "text-gray-200 font-semibold ml-2 mr-1",
-						},
-						elem.Text(n.ID),
-					),
-					elem.Text("Status: "),
-					elem.If(
-						n.IsOnline(),
-						elem.I(
-							attrs.Props{
-								"class": "fas fa-circle animate-pulse text-green-500 ml-2 mr-1",
-							},
-						),
-						elem.I(
-							attrs.Props{
-								"class": "fas fa-circle animate-pulse text-red-500 ml-2 mr-1",
-							},
-						),
-					),
-					elem.If(
-						n.IsOnline(),
-						elem.Span(
-							attrs.Props{
-								"class": "text-green-400",
-							},
-
-							elem.Text("Online"),
-						),
-						elem.Span(
-							attrs.Props{
-								"class": "text-red-400",
-							},
-							elem.Text("Offline"),
-						),
-					),
-				),
-			))
-	}
-
-	return renderElements(nodesElements)
-}
-
 func StartProgressBar(uid, progress, text string) string {
 	if progress == "" {
 		progress = "0"

core/http/endpoints/elevenlabs/tts.go

@@ -28,7 +28,7 @@ func TTSEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfi
 			return err
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.ModelID, false)
+		modelFile, err := fiberContext.ModelFromContext(c, ml, input.ModelID, false)
 		if err != nil {
 			modelFile = input.ModelID
 			log.Warn().Msgf("Model not found in context: %s", input.ModelID)

core/http/endpoints/jina/rerank.go

@@ -12,11 +12,6 @@ import (
 	"github.com/rs/zerolog/log"
 )
 
-// JINARerankEndpoint acts like the Jina reranker endpoint (https://jina.ai/reranker/)
-// @Summary Reranks a list of phrases by relevance to a given text query.
-// @Param request body schema.JINARerankRequest true "query params"
-// @Success 200 {object} schema.JINARerankResponse "Response"
-// @Router /v1/rerank [post]
 func JINARerankEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		req := new(schema.JINARerankRequest)
@@ -33,7 +28,7 @@ func JINARerankEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, a
 			return err
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, false)
+		modelFile, err := fiberContext.ModelFromContext(c, ml, input.Model, false)
 		if err != nil {
 			modelFile = input.Model
 			log.Warn().Msgf("Model not found in context: %s", input.Model)

core/http/endpoints/localai/backend_monitor.go

@@ -6,11 +6,6 @@ import (
 	"github.com/mudler/LocalAI/core/services"
 )
 
-// BackendMonitorEndpoint returns the status of the specified backend
-// @Summary Backend monitor endpoint
-// @Param request body schema.BackendMonitorRequest true "Backend statistics request"
-// @Success 200 {object} proto.StatusResponse "Response"
-// @Router /backend/monitor [get]
 func BackendMonitorEndpoint(bm *services.BackendMonitorService) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 
@@ -28,10 +23,6 @@ func BackendMonitorEndpoint(bm *services.BackendMonitorService) func(c *fiber.Ct
 	}
 }
 
-// BackendMonitorEndpoint shuts down the specified backend
-// @Summary Backend monitor endpoint
-// @Param request body schema.BackendMonitorRequest true "Backend statistics request"
-// @Router /backend/shutdown [post]
 func BackendShutdownEndpoint(bm *services.BackendMonitorService) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(schema.BackendMonitorRequest)

core/http/endpoints/localai/gallery.go

@@ -9,7 +9,6 @@ import (
 	"github.com/google/uuid"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/gallery"
-	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/rs/zerolog/log"
 )
@@ -34,10 +33,6 @@ func CreateModelGalleryEndpointService(galleries []config.Gallery, modelPath str
 	}
 }
 
-// GetOpStatusEndpoint returns the job status
-// @Summary Returns the job status
-// @Success 200 {object} gallery.GalleryOpStatus "Response"
-// @Router /models/jobs/{uuid} [get]
 func (mgs *ModelGalleryEndpointService) GetOpStatusEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		status := mgs.galleryApplier.GetStatus(c.Params("uuid"))
@@ -48,21 +43,12 @@ func (mgs *ModelGalleryEndpointService) GetOpStatusEndpoint() func(c *fiber.Ctx)
 	}
 }
 
-// GetAllStatusEndpoint returns all the jobs status progress
-// @Summary Returns all the jobs status progress
-// @Success 200 {object} map[string]gallery.GalleryOpStatus "Response"
-// @Router /models/jobs [get]
 func (mgs *ModelGalleryEndpointService) GetAllStatusEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		return c.JSON(mgs.galleryApplier.GetAllStatus())
 	}
 }
 
-// ApplyModelGalleryEndpoint installs a new model to a LocalAI instance from the model gallery
-// @Summary Install models to LocalAI.
-// @Param request body GalleryModel true "query params"
-// @Success 200 {object} schema.GalleryResponse "Response"
-// @Router /models/apply [post]
 func (mgs *ModelGalleryEndpointService) ApplyModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(GalleryModel)
@@ -82,15 +68,13 @@ func (mgs *ModelGalleryEndpointService) ApplyModelGalleryEndpoint() func(c *fibe
 			Galleries:        mgs.galleries,
 			ConfigURL:        input.ConfigURL,
 		}
-		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
+		return c.JSON(struct {
+			ID        string `json:"uuid"`
+			StatusURL string `json:"status"`
+		}{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
 	}
 }
 
-// DeleteModelGalleryEndpoint lets delete models from a LocalAI instance
-// @Summary delete models to LocalAI.
-// @Param name	path string	true	"Model name"
-// @Success 200 {object} schema.GalleryResponse "Response"
-// @Router /models/delete/{name} [post]
 func (mgs *ModelGalleryEndpointService) DeleteModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		modelName := c.Params("name")
@@ -105,14 +89,13 @@ func (mgs *ModelGalleryEndpointService) DeleteModelGalleryEndpoint() func(c *fib
 			return err
 		}
 
-		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
+		return c.JSON(struct {
+			ID        string `json:"uuid"`
+			StatusURL string `json:"status"`
+		}{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
 	}
 }
 
-// ListModelFromGalleryEndpoint list the available models for installation from the active galleries
-// @Summary List installable models.
-// @Success 200 {object} []gallery.GalleryModel "Response"
-// @Router /models/available [get]
 func (mgs *ModelGalleryEndpointService) ListModelFromGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		log.Debug().Msgf("Listing models from galleries: %+v", mgs.galleries)
@@ -133,10 +116,6 @@ func (mgs *ModelGalleryEndpointService) ListModelFromGalleryEndpoint() func(c *f
 	}
 }
 
-// ListModelGalleriesEndpoint list the available galleries configured in LocalAI
-// @Summary List all Galleries
-// @Success 200 {object} []config.Gallery "Response"
-// @Router /models/galleries [get]
 // NOTE: This is different (and much simpler!) than above! This JUST lists the model galleries that have been loaded, not their contents!
 func (mgs *ModelGalleryEndpointService) ListModelGalleriesEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
@@ -149,11 +128,6 @@ func (mgs *ModelGalleryEndpointService) ListModelGalleriesEndpoint() func(c *fib
 	}
 }
 
-// AddModelGalleryEndpoint adds a gallery in LocalAI
-// @Summary Adds a gallery in LocalAI
-// @Param request body config.Gallery true "Gallery details"
-// @Success 200 {object} []config.Gallery "Response"
-// @Router /models/galleries [post]
 func (mgs *ModelGalleryEndpointService) AddModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(config.Gallery)
@@ -176,11 +150,6 @@ func (mgs *ModelGalleryEndpointService) AddModelGalleryEndpoint() func(c *fiber.
 	}
 }
 
-// RemoveModelGalleryEndpoint remove a gallery in LocalAI
-// @Summary removes a gallery from LocalAI
-// @Param request body config.Gallery true "Gallery details"
-// @Success 200 {object} []config.Gallery "Response"
-// @Router /models/galleries [delete]
 func (mgs *ModelGalleryEndpointService) RemoveModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(config.Gallery)
@@ -196,10 +165,6 @@ func (mgs *ModelGalleryEndpointService) RemoveModelGalleryEndpoint() func(c *fib
 		mgs.galleries = slices.DeleteFunc(mgs.galleries, func(gallery config.Gallery) bool {
 			return gallery.Name == input.Name
 		})
-		dat, err := json.Marshal(mgs.galleries)
-		if err != nil {
-			return err
-		}
-		return c.Send(dat)
+		return c.Send(nil)
 	}
 }

core/http/endpoints/localai/metrics.go

@@ -9,11 +9,8 @@ import (
 	"github.com/prometheus/client_golang/prometheus/promhttp"
 )
 
-// LocalAIMetricsEndpoint returns the metrics endpoint for LocalAI
-// @Summary Prometheus metrics endpoint
-// @Param request body config.Gallery true "Gallery details"
-// @Router /metrics [get]
 func LocalAIMetricsEndpoint() fiber.Handler {
+
 	return adaptor.HTTPHandler(promhttp.Handler())
 }
 

core/http/endpoints/localai/p2p.go

@@ -1,28 +0,0 @@
-package localai
-
-import (
-	"github.com/gofiber/fiber/v2"
-	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/p2p"
-	"github.com/mudler/LocalAI/core/schema"
-)
-
-// ShowP2PNodes returns the P2P Nodes
-// @Summary Returns available P2P nodes
-// @Success 200 {object} []schema.P2PNodesResponse "Response"
-// @Router /api/p2p [get]
-func ShowP2PNodes(c *fiber.Ctx) error {
-	// Render index
-	return c.JSON(schema.P2PNodesResponse{
-		Nodes:          p2p.GetAvailableNodes(""),
-		FederatedNodes: p2p.GetAvailableNodes(p2p.FederatedID),
-	})
-}
-
-// ShowP2PToken returns the P2P token
-// @Summary Show the P2P token
-// @Success 200 {string} string	 "Response"
-// @Router /api/p2p/token [get]
-func ShowP2PToken(appConfig *config.ApplicationConfig) func(*fiber.Ctx) error {
-	return func(c *fiber.Ctx) error { return c.Send([]byte(appConfig.P2PToken)) }
-}

core/http/endpoints/localai/tts.go

@@ -29,7 +29,7 @@ func TTSEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfi
 			return err
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, false)
+		modelFile, err := fiberContext.ModelFromContext(c, ml, input.Model, false)
 		if err != nil {
 			modelFile = input.Model
 			log.Warn().Msgf("Model not found in context: %s", input.Model)

core/http/endpoints/localai/welcome.go

@@ -4,8 +4,6 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/gallery"
-	"github.com/mudler/LocalAI/core/p2p"
-	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/internal"
 	"github.com/mudler/LocalAI/pkg/model"
 )
@@ -13,7 +11,7 @@ import (
 func WelcomeEndpoint(appConfig *config.ApplicationConfig,
 	cl *config.BackendConfigLoader, ml *model.ModelLoader, modelStatus func() (map[string]string, map[string]string)) func(*fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		models, _ := services.ListModels(cl, ml, "", true)
+		models, _ := ml.ListModels()
 		backendConfigs := cl.GetAllBackendConfigs()
 
 		galleryConfigs := map[string]*gallery.Config{}
@@ -29,21 +27,12 @@ func WelcomeEndpoint(appConfig *config.ApplicationConfig,
 		// Get model statuses to display in the UI the operation in progress
 		processingModels, taskTypes := modelStatus()
 
-		modelsWithoutConfig := []string{}
-
-		for _, m := range models {
-			if _, ok := galleryConfigs[m]; !ok {
-				modelsWithoutConfig = append(modelsWithoutConfig, m)
-			}
-		}
-
 		summary := fiber.Map{
 			"Title":             "LocalAI API - " + internal.PrintableVersion(),
 			"Version":           internal.PrintableVersion(),
-			"Models":            modelsWithoutConfig,
+			"Models":            models,
 			"ModelsConfig":      backendConfigs,
 			"GalleryConfig":     galleryConfigs,
-			"IsP2PEnabled":      p2p.IsP2PEnabled(),
 			"ApplicationConfig": appConfig,
 			"ProcessingModels":  processingModels,
 			"TaskTypes":         taskTypes,

core/http/endpoints/openai/assistant.go

@@ -11,8 +11,6 @@ import (
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/schema"
-	"github.com/mudler/LocalAI/core/services"
 	model "github.com/mudler/LocalAI/pkg/model"
 	"github.com/mudler/LocalAI/pkg/utils"
 	"github.com/rs/zerolog/log"
@@ -81,7 +79,7 @@ func CreateAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoad
 			return c.Status(fiber.StatusBadRequest).JSON(fiber.Map{"error": "Cannot parse JSON"})
 		}
 
-		if !modelExists(cl, ml, request.Model) {
+		if !modelExists(ml, request.Model) {
 			log.Warn().Msgf("Model: %s was not found in list of models.", request.Model)
 			return c.Status(fiber.StatusBadRequest).SendString("Model " + request.Model + " not found")
 		}
@@ -126,14 +124,6 @@ func generateRandomID() int64 {
 	return currentId
 }
 
-// ListAssistantsEndpoint is the OpenAI Assistant API endpoint to list assistents https://platform.openai.com/docs/api-reference/assistants/listAssistants
-// @Summary List available assistents
-// @Param limit query int false "Limit the number of assistants returned"
-// @Param order query string false "Order of assistants returned"
-// @Param after query string false "Return assistants created after the given ID"
-// @Param before query string false "Return assistants created before the given ID"
-// @Success 200 {object} []Assistant "Response"
-// @Router /v1/assistants [get]
 func ListAssistantsEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		// Because we're altering the existing assistants list we should just duplicate it for now.
@@ -223,9 +213,9 @@ func filterAssistantsAfterID(assistants []Assistant, id string) []Assistant {
 	return filteredAssistants
 }
 
-func modelExists(cl *config.BackendConfigLoader, ml *model.ModelLoader, modelName string) (found bool) {
+func modelExists(ml *model.ModelLoader, modelName string) (found bool) {
 	found = false
-	models, err := services.ListModels(cl, ml, "", true)
+	models, err := ml.ListModels()
 	if err != nil {
 		return
 	}
@@ -239,11 +229,13 @@ func modelExists(cl *config.BackendConfigLoader, ml *model.ModelLoader, modelNam
 	return
 }
 
-// DeleteAssistantEndpoint is the OpenAI Assistant API endpoint to delete assistents https://platform.openai.com/docs/api-reference/assistants/deleteAssistant
-// @Summary Delete assistents
-// @Success 200 {object} schema.DeleteAssistantResponse "Response"
-// @Router /v1/assistants/{assistant_id} [delete]
 func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
+	type DeleteAssistantResponse struct {
+		ID      string `json:"id"`
+		Object  string `json:"object"`
+		Deleted bool   `json:"deleted"`
+	}
+
 	return func(c *fiber.Ctx) error {
 		assistantID := c.Params("assistant_id")
 		if assistantID == "" {
@@ -254,7 +246,7 @@ func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoad
 			if assistant.ID == assistantID {
 				Assistants = append(Assistants[:i], Assistants[i+1:]...)
 				utils.SaveConfig(appConfig.ConfigsDir, AssistantsConfigFile, Assistants)
-				return c.Status(fiber.StatusOK).JSON(schema.DeleteAssistantResponse{
+				return c.Status(fiber.StatusOK).JSON(DeleteAssistantResponse{
 					ID:      assistantID,
 					Object:  "assistant.deleted",
 					Deleted: true,
@@ -263,7 +255,7 @@ func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoad
 		}
 
 		log.Warn().Msgf("Unable to find assistant %s for deletion", assistantID)
-		return c.Status(fiber.StatusNotFound).JSON(schema.DeleteAssistantResponse{
+		return c.Status(fiber.StatusNotFound).JSON(DeleteAssistantResponse{
 			ID:      assistantID,
 			Object:  "assistant.deleted",
 			Deleted: false,
@@ -271,10 +263,6 @@ func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoad
 	}
 }
 
-// GetAssistantEndpoint is the OpenAI Assistant API endpoint to get assistents https://platform.openai.com/docs/api-reference/assistants/getAssistant
-// @Summary Get assistent data
-// @Success 200 {object} Assistant "Response"
-// @Router /v1/assistants/{assistant_id} [get]
 func GetAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		assistantID := c.Params("assistant_id")
@@ -304,9 +292,19 @@ var (
 	AssistantsFileConfigFile = "assistantsFile.json"
 )
 
+type AssistantFileRequest struct {
+	FileID string `json:"file_id"`
+}
+
+type DeleteAssistantFileResponse struct {
+	ID      string `json:"id"`
+	Object  string `json:"object"`
+	Deleted bool   `json:"deleted"`
+}
+
 func CreateAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		request := new(schema.AssistantFileRequest)
+		request := new(AssistantFileRequest)
 		if err := c.BodyParser(request); err != nil {
 			return c.Status(fiber.StatusBadRequest).JSON(fiber.Map{"error": "Cannot parse JSON"})
 		}
@@ -347,7 +345,7 @@ func CreateAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 
 func ListAssistantFilesEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	type ListAssistantFiles struct {
-		Data   []schema.File
+		Data   []File
 		Object string
 	}
 
@@ -465,7 +463,7 @@ func DeleteAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 							// Remove the file from the assistantFiles slice
 							AssistantFiles = append(AssistantFiles[:i], AssistantFiles[i+1:]...)
 							utils.SaveConfig(appConfig.ConfigsDir, AssistantsFileConfigFile, AssistantFiles)
-							return c.Status(fiber.StatusOK).JSON(schema.DeleteAssistantFileResponse{
+							return c.Status(fiber.StatusOK).JSON(DeleteAssistantFileResponse{
 								ID:      fileId,
 								Object:  "assistant.file.deleted",
 								Deleted: true,
@@ -481,7 +479,7 @@ func DeleteAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 						AssistantFiles = append(AssistantFiles[:i], AssistantFiles[i+1:]...)
 						utils.SaveConfig(appConfig.ConfigsDir, AssistantsFileConfigFile, AssistantFiles)
 
-						return c.Status(fiber.StatusNotFound).JSON(schema.DeleteAssistantFileResponse{
+						return c.Status(fiber.StatusNotFound).JSON(DeleteAssistantFileResponse{
 							ID:      fileId,
 							Object:  "assistant.file.deleted",
 							Deleted: true,
@@ -492,7 +490,7 @@ func DeleteAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 		}
 		log.Warn().Msgf("Unable to find assistant: %s", assistantID)
 
-		return c.Status(fiber.StatusNotFound).JSON(schema.DeleteAssistantFileResponse{
+		return c.Status(fiber.StatusNotFound).JSON(DeleteAssistantFileResponse{
 			ID:      fileId,
 			Object:  "assistant.file.deleted",
 			Deleted: false,

core/http/endpoints/openai/assistant_test.go

@@ -14,7 +14,6 @@ import (
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/stretchr/testify/assert"
 )
@@ -27,7 +26,7 @@ type MockLoader struct {
 
 func tearDown() func() {
 	return func() {
-		UploadedFiles = []schema.File{}
+		UploadedFiles = []File{}
 		Assistants = []Assistant{}
 		AssistantFiles = []AssistantFile{}
 		_ = os.Remove(filepath.Join(configsDir, AssistantsConfigFile))
@@ -295,7 +294,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)
 
-		afr := schema.AssistantFileRequest{FileID: file.ID}
+		afr := AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)
 
 		assert.NoError(t, err)
@@ -306,7 +305,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)
 
-		afr := schema.AssistantFileRequest{FileID: file.ID}
+		afr := AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)
 		assert.NoError(t, err)
 
@@ -317,7 +316,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)
 
-		afr := schema.AssistantFileRequest{FileID: file.ID}
+		afr := AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)
 		assert.NoError(t, err)
 		t.Cleanup(cleanupAssistantFile(t, app, af.ID, af.AssistantID))
@@ -339,7 +338,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)
 
-		afr := schema.AssistantFileRequest{FileID: file.ID}
+		afr := AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)
 		assert.NoError(t, err)
 
@@ -350,7 +349,7 @@ func TestAssistantEndpoints(t *testing.T) {
 
 }
 
-func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationConfig) (schema.File, Assistant, error) {
+func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationConfig) (File, Assistant, error) {
 	ar := &AssistantRequest{
 		Model:        "ggml-gpt4all-j",
 		Name:         "3.5-turbo",
@@ -363,7 +362,7 @@ func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationC
 
 	assistant, _, err := createAssistant(app, *ar)
 	if err != nil {
-		return schema.File{}, Assistant{}, err
+		return File{}, Assistant{}, err
 	}
 	t.Cleanup(cleanupAllAssistants(t, app, []string{assistant.ID}))
 
@@ -375,7 +374,7 @@ func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationC
 	return file, assistant, nil
 }
 
-func createAssistantFile(app *fiber.App, afr schema.AssistantFileRequest, assistantId string) (AssistantFile, *http.Response, error) {
+func createAssistantFile(app *fiber.App, afr AssistantFileRequest, assistantId string) (AssistantFile, *http.Response, error) {
 	afrJson, err := json.Marshal(afr)
 	if err != nil {
 		return AssistantFile{}, nil, err
@@ -452,7 +451,7 @@ func cleanupAssistantFile(t *testing.T, app *fiber.App, fileId, assistantId stri
 		resp, err := app.Test(request)
 		assert.NoError(t, err)
 
-		var dafr schema.DeleteAssistantFileResponse
+		var dafr DeleteAssistantFileResponse
 		err = json.NewDecoder(resp.Body).Decode(&dafr)
 		assert.NoError(t, err)
 		assert.True(t, dafr.Deleted)

core/http/endpoints/openai/chat.go

@@ -159,7 +159,7 @@ func ChatEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, startup
 	}
 
 	return func(c *fiber.Ctx) error {
-		modelFile, input, err := readRequest(c, cl, ml, startupOptions, true)
+		modelFile, input, err := readRequest(c, ml, startupOptions, true)
 		if err != nil {
 			return fmt.Errorf("failed reading parameters from request:%w", err)
 		}
@@ -225,10 +225,18 @@ func ChatEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, startup
 			}
 
 			// Update input grammar
-			jsStruct := funcs.ToJSONStructure(config.FunctionsConfig.FunctionNameKey, config.FunctionsConfig.FunctionNameKey)
-			config.Grammar = jsStruct.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+			// Handle if we should return "name" instead of "functions"
+			if config.FunctionsConfig.FunctionName {
+				jsStruct := funcs.ToJSONNameStructure()
+				config.Grammar = jsStruct.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+			} else {
+				jsStruct := funcs.ToJSONFunctionStructure()
+				config.Grammar = jsStruct.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+			}
 		case input.JSONFunctionGrammarObject != nil:
 			config.Grammar = input.JSONFunctionGrammarObject.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
+		case input.JSONFunctionGrammarObjectName != nil:
+			config.Grammar = input.JSONFunctionGrammarObjectName.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
 		default:
 			// Force picking one of the functions by the request
 			if config.FunctionToCall() != "" {

core/http/endpoints/openai/completion.go

@@ -57,7 +57,7 @@ func CompletionEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, a
 	}
 
 	return func(c *fiber.Ctx) error {
-		modelFile, input, err := readRequest(c, cl, ml, appConfig, true)
+		modelFile, input, err := readRequest(c, ml, appConfig, true)
 		if err != nil {
 			return fmt.Errorf("failed reading parameters from request:%w", err)
 		}

core/http/endpoints/openai/edit.go

@@ -16,14 +16,9 @@ import (
 	"github.com/rs/zerolog/log"
 )
 
-// EditEndpoint is the OpenAI edit API endpoint
-// @Summary OpenAI edit endpoint
-// @Param request body schema.OpenAIRequest true "query params"
-// @Success 200 {object} schema.OpenAIResponse "Response"
-// @Router /v1/edits [post]
 func EditEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		modelFile, input, err := readRequest(c, cl, ml, appConfig, true)
+		modelFile, input, err := readRequest(c, ml, appConfig, true)
 		if err != nil {
 			return fmt.Errorf("failed reading parameters from request:%w", err)
 		}

core/http/endpoints/openai/embeddings.go

@@ -23,7 +23,7 @@ import (
 // @Router /v1/embeddings [post]
 func EmbeddingsEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		model, input, err := readRequest(c, cl, ml, appConfig, true)
+		model, input, err := readRequest(c, ml, appConfig, true)
 		if err != nil {
 			return fmt.Errorf("failed reading parameters from request:%w", err)
 		}

core/http/endpoints/openai/files.go

@@ -9,16 +9,25 @@ import (
 	"time"
 
 	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/schema"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/pkg/utils"
 )
 
-var UploadedFiles []schema.File
+var UploadedFiles []File
 
 const UploadedFilesFile = "uploadedFiles.json"
 
+// File represents the structure of a file object from the OpenAI API.
+type File struct {
+	ID        string    `json:"id"`         // Unique identifier for the file
+	Object    string    `json:"object"`     // Type of the object (e.g., "file")
+	Bytes     int       `json:"bytes"`      // Size of the file in bytes
+	CreatedAt time.Time `json:"created_at"` // The time at which the file was created
+	Filename  string    `json:"filename"`   // The name of the file
+	Purpose   string    `json:"purpose"`    // The purpose of the file (e.g., "fine-tune", "classifications", etc.)
+}
+
 // UploadFilesEndpoint https://platform.openai.com/docs/api-reference/files/create
 func UploadFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
@@ -52,7 +61,7 @@ func UploadFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Appli
 			return c.Status(fiber.StatusInternalServerError).SendString("Failed to save file: " + err.Error())
 		}
 
-		f := schema.File{
+		f := File{
 			ID:        fmt.Sprintf("file-%d", getNextFileId()),
 			Object:    "file",
 			Bytes:     int(file.Size),
@@ -75,13 +84,14 @@ func getNextFileId() int64 {
 }
 
 // ListFilesEndpoint https://platform.openai.com/docs/api-reference/files/list
-// @Summary List files.
-// @Success 200 {object} schema.ListFiles "Response"
-// @Router /v1/files [get]
 func ListFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
+	type ListFiles struct {
+		Data   []File
+		Object string
+	}
 
 	return func(c *fiber.Ctx) error {
-		var listFiles schema.ListFiles
+		var listFiles ListFiles
 
 		purpose := c.Query("purpose")
 		if purpose == "" {
@@ -98,7 +108,7 @@ func ListFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Applica
 	}
 }
 
-func getFileFromRequest(c *fiber.Ctx) (*schema.File, error) {
+func getFileFromRequest(c *fiber.Ctx) (*File, error) {
 	id := c.Params("file_id")
 	if id == "" {
 		return nil, fmt.Errorf("file_id parameter is required")
@@ -113,10 +123,7 @@ func getFileFromRequest(c *fiber.Ctx) (*schema.File, error) {
 	return nil, fmt.Errorf("unable to find file id %s", id)
 }
 
-// GetFilesEndpoint is the OpenAI API endpoint to get files https://platform.openai.com/docs/api-reference/files/retrieve
-// @Summary Returns information about a specific file.
-// @Success 200 {object} schema.File "Response"
-// @Router /v1/files/{file_id} [get]
+// GetFilesEndpoint https://platform.openai.com/docs/api-reference/files/retrieve
 func GetFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		file, err := getFileFromRequest(c)
@@ -128,17 +135,13 @@ func GetFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Applicat
 	}
 }
 
-type DeleteStatus struct {
-	Id      string
-	Object  string
-	Deleted bool
-}
-
-// DeleteFilesEndpoint is the OpenAI API endpoint to delete files https://platform.openai.com/docs/api-reference/files/delete
-// @Summary Delete a file.
-// @Success 200 {object} DeleteStatus "Response"
-// @Router /v1/files/{file_id} [delete]
+// DeleteFilesEndpoint https://platform.openai.com/docs/api-reference/files/delete
 func DeleteFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
+	type DeleteStatus struct {
+		Id      string
+		Object  string
+		Deleted bool
+	}
 
 	return func(c *fiber.Ctx) error {
 		file, err := getFileFromRequest(c)
@@ -171,11 +174,7 @@ func DeleteFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Appli
 	}
 }
 
-// GetFilesContentsEndpoint is the OpenAI API endpoint to get files content https://platform.openai.com/docs/api-reference/files/retrieve-contents
-// @Summary Returns information about a specific file.
-// @Success	200		{string}	binary				"file"
-// @Router /v1/files/{file_id}/content [get]
-// GetFilesContentsEndpoint
+// GetFilesContentsEndpoint https://platform.openai.com/docs/api-reference/files/retrieve-contents
 func GetFilesContentsEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		file, err := getFileFromRequest(c)

core/http/endpoints/openai/files_test.go

@@ -14,7 +14,6 @@ import (
 	"github.com/rs/zerolog/log"
 
 	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/schema"
 
 	"github.com/gofiber/fiber/v2"
 	utils2 "github.com/mudler/LocalAI/pkg/utils"
@@ -23,6 +22,11 @@ import (
 	"testing"
 )
 
+type ListFiles struct {
+	Data   []File
+	Object string
+}
+
 func startUpApp() (app *fiber.App, option *config.ApplicationConfig, loader *config.BackendConfigLoader) {
 	// Preparing the mocked objects
 	loader = &config.BackendConfigLoader{}
@@ -155,7 +159,7 @@ func TestUploadFileExceedSizeLimit(t *testing.T) {
 		resp, _ := app.Test(req)
 		assert.Equal(t, 200, resp.StatusCode)
 
-		var listFiles schema.ListFiles
+		var listFiles ListFiles
 		if err := json.Unmarshal(bodyToByteArray(resp, t), &listFiles); err != nil {
 			t.Errorf("Failed to decode response: %v", err)
 			return
@@ -197,7 +201,7 @@ func CallFilesUploadEndpoint(t *testing.T, app *fiber.App, fileName, tag, purpos
 	return app.Test(req)
 }
 
-func CallFilesUploadEndpointWithCleanup(t *testing.T, app *fiber.App, fileName, tag, purpose string, fileSize int, appConfig *config.ApplicationConfig) schema.File {
+func CallFilesUploadEndpointWithCleanup(t *testing.T, app *fiber.App, fileName, tag, purpose string, fileSize int, appConfig *config.ApplicationConfig) File {
 	// Create a file that exceeds the limit
 	testName := strings.Split(t.Name(), "/")[1]
 	file := createTestFile(t, testName+"-"+fileName, fileSize, appConfig)
@@ -276,8 +280,8 @@ func bodyToByteArray(resp *http.Response, t *testing.T) []byte {
 	return bodyBytes
 }
 
-func responseToFile(t *testing.T, resp *http.Response) schema.File {
-	var file schema.File
+func responseToFile(t *testing.T, resp *http.Response) File {
+	var file File
 	responseToString := bodyToString(resp, t)
 
 	err := json.NewDecoder(strings.NewReader(responseToString)).Decode(&file)
@@ -288,8 +292,8 @@ func responseToFile(t *testing.T, resp *http.Response) schema.File {
 	return file
 }
 
-func responseToListFile(t *testing.T, resp *http.Response) schema.ListFiles {
-	var listFiles schema.ListFiles
+func responseToListFile(t *testing.T, resp *http.Response) ListFiles {
+	var listFiles ListFiles
 	responseToString := bodyToString(resp, t)
 
 	err := json.NewDecoder(strings.NewReader(responseToString)).Decode(&listFiles)

core/http/endpoints/openai/image.go

@@ -66,7 +66,7 @@ func downloadFile(url string) (string, error) {
 // @Router /v1/images/generations [post]
 func ImageEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		m, input, err := readRequest(c, cl, ml, appConfig, false)
+		m, input, err := readRequest(c, ml, appConfig, false)
 		if err != nil {
 			return fmt.Errorf("failed reading parameters from request:%w", err)
 		}

core/http/endpoints/openai/list.go

@@ -2,17 +2,11 @@ package openai
 
 import (
 	"github.com/gofiber/fiber/v2"
-	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/core/services"
-	model "github.com/mudler/LocalAI/pkg/model"
 )
 
-// ListModelsEndpoint is the OpenAI Models API endpoint https://platform.openai.com/docs/api-reference/models
-// @Summary List and describe the various models available in the API.
-// @Success 200 {object} schema.ModelsDataResponse "Response"
-// @Router /v1/models [get]
-func ListModelsEndpoint(bcl *config.BackendConfigLoader, ml *model.ModelLoader) func(ctx *fiber.Ctx) error {
+func ListModelsEndpoint(lms *services.ListModelsService) func(ctx *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		// If blank, no filter is applied.
 		filter := c.Query("filter")
@@ -20,30 +14,16 @@ func ListModelsEndpoint(bcl *config.BackendConfigLoader, ml *model.ModelLoader)
 		// By default, exclude any loose files that are already referenced by a configuration file.
 		excludeConfigured := c.QueryBool("excludeConfigured", true)
 
-		dataModels, err := modelList(bcl, ml, filter, excludeConfigured)
+		dataModels, err := lms.ListModels(filter, excludeConfigured)
 		if err != nil {
 			return err
 		}
-		return c.JSON(schema.ModelsDataResponse{
+		return c.JSON(struct {
+			Object string               `json:"object"`
+			Data   []schema.OpenAIModel `json:"data"`
+		}{
 			Object: "list",
 			Data:   dataModels,
 		})
 	}
 }
-
-func modelList(bcl *config.BackendConfigLoader, ml *model.ModelLoader, filter string, excludeConfigured bool) ([]schema.OpenAIModel, error) {
-
-	models, err := services.ListModels(bcl, ml, filter, excludeConfigured)
-	if err != nil {
-		return nil, err
-	}
-
-	dataModels := []schema.OpenAIModel{}
-
-	// Then iterate through the loose files:
-	for _, m := range models {
-		dataModels = append(dataModels, schema.OpenAIModel{ID: m, Object: "model"})
-	}
-
-	return dataModels, nil
-}

core/http/endpoints/openai/request.go

@@ -15,7 +15,7 @@ import (
 	"github.com/rs/zerolog/log"
 )
 
-func readRequest(c *fiber.Ctx, cl *config.BackendConfigLoader, ml *model.ModelLoader, o *config.ApplicationConfig, firstModel bool) (string, *schema.OpenAIRequest, error) {
+func readRequest(c *fiber.Ctx, ml *model.ModelLoader, o *config.ApplicationConfig, firstModel bool) (string, *schema.OpenAIRequest, error) {
 	input := new(schema.OpenAIRequest)
 
 	// Get input data from the request body
@@ -31,7 +31,7 @@ func readRequest(c *fiber.Ctx, cl *config.BackendConfigLoader, ml *model.ModelLo
 
 	log.Debug().Msgf("Request received: %s", string(received))
 
-	modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, firstModel)
+	modelFile, err := fiberContext.ModelFromContext(c, ml, input.Model, firstModel)
 
 	return modelFile, input, err
 }

core/http/endpoints/openai/transcription.go

@@ -25,7 +25,7 @@ import (
 // @Router /v1/audio/transcriptions [post]
 func TranscriptEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		m, input, err := readRequest(c, cl, ml, appConfig, false)
+		m, input, err := readRequest(c, ml, appConfig, false)
 		if err != nil {
 			return fmt.Errorf("failed reading parameters from request:%w", err)
 		}

core/http/routes/localai.go

@@ -5,7 +5,6 @@ import (
 	"github.com/gofiber/swagger"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/http/endpoints/localai"
-	"github.com/mudler/LocalAI/core/p2p"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/internal"
 	"github.com/mudler/LocalAI/pkg/model"
@@ -57,12 +56,6 @@ func RegisterLocalAIRoutes(app *fiber.App,
 	app.Get("/backend/monitor", auth, localai.BackendMonitorEndpoint(backendMonitorService))
 	app.Post("/backend/shutdown", auth, localai.BackendShutdownEndpoint(backendMonitorService))
 
-	// p2p
-	if p2p.IsP2PEnabled() {
-		app.Get("/api/p2p", auth, localai.ShowP2PNodes)
-		app.Get("/api/p2p/token", auth, localai.ShowP2PToken(appConfig))
-	}
-
 	app.Get("/version", auth, func(c *fiber.Ctx) error {
 		return c.JSON(struct {
 			Version string `json:"version"`

core/http/routes/openai.go

@@ -5,6 +5,7 @@ import (
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/http/endpoints/localai"
 	"github.com/mudler/LocalAI/core/http/endpoints/openai"
+	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/pkg/model"
 )
 
@@ -80,7 +81,8 @@ func RegisterOpenAIRoutes(app *fiber.App,
 		app.Static("/generated-audio", appConfig.AudioDir)
 	}
 
-	// List models
-	app.Get("/v1/models", auth, openai.ListModelsEndpoint(cl, ml))
-	app.Get("/models", auth, openai.ListModelsEndpoint(cl, ml))
+	// models
+	tmpLMS := services.NewListModelsService(ml, cl, appConfig) // TODO: once createApplication() is fully in use, reference the central instance.
+	app.Get("/v1/models", auth, openai.ListModelsEndpoint(tmpLMS))
+	app.Get("/models", auth, openai.ListModelsEndpoint(tmpLMS))
 }