chore(deps): bump llama.cpp to '10f2e81809bbb69ecfe64fc8b4686285f84b0c07'

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

.env

@@ -29,9 +29,6 @@
 ## Enable/Disable single backend (useful if only one GPU is available)
 # LOCALAI_SINGLE_ACTIVE_BACKEND=true
 
-# Forces shutdown of the backends if busy (only if LOCALAI_SINGLE_ACTIVE_BACKEND is set)
-# LOCALAI_FORCE_BACKEND_SHUTDOWN=true
-
 ## Specify a build type. Available: cublas, openblas, clblas.
 ## cuBLAS: This is a GPU-accelerated version of the complete standard BLAS (Basic Linear Algebra Subprograms) library. It's provided by Nvidia and is part of their CUDA toolkit.
 ## OpenBLAS: This is an open-source implementation of the BLAS library that aims to provide highly optimized code for various platforms. It includes support for multi-threading and can be compiled to use hardware-specific features for additional performance. OpenBLAS can run on many kinds of hardware, including CPUs from Intel, AMD, and ARM.
@@ -76,7 +73,7 @@
 
 ### Define a list of GRPC Servers for llama-cpp workers to distribute the load
 # https://github.com/ggerganov/llama.cpp/pull/6829
-# https://github.com/ggerganov/llama.cpp/blob/master/tools/rpc/README.md
+# https://github.com/ggerganov/llama.cpp/blob/master/examples/rpc/README.md
 # LLAMACPP_GRPC_SERVERS=""
 
 ### Enable to run parallel requests

.github/dependabot.yml

@@ -29,6 +29,10 @@ 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:
@@ -61,6 +65,10 @@ updates:
     directory: "/backend/python/openvoice"
     schedule:
       interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/parler-tts"
+    schedule:
+      interval: "weekly"
   - package-ecosystem: "pip"
     directory: "/backend/python/rerankers"
     schedule:

.github/workflows/backend.yml

@@ -1,554 +0,0 @@
----
-name: 'build backend container images'
-
-on:
-  push:
-    branches:
-      - master
-    tags:
-      - '*'
-  #pull_request:
-
-concurrency:
-  group: ci-backends-${{ github.head_ref || github.ref }}-${{ github.repository }}
-  cancel-in-progress: true
-
-jobs:
-  backend-jobs:
-    uses: ./.github/workflows/backend_build.yml
-    with:
-      tag-latest: ${{ matrix.tag-latest }}
-      tag-suffix: ${{ matrix.tag-suffix }}
-      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 }}
-      backend: ${{ matrix.backend }}
-      latest-image: ${{ matrix.latest-image }}
-      dockerfile: $${ matrix.dockerfile }}
-      context: $${ matrix.context }}
-    secrets:
-      dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
-      dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
-      quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
-      quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
-    strategy:
-      fail-fast: false
-      #max-parallel: ${{ github.event_name != 'pull_request' && 6 || 4 }}
-      matrix:
-        include:
-          # CUDA 11 builds
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-rerankers'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "rerankers"
-            latest-image: 'latest-gpu-nvidia-cuda-11-rerankers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-vllm'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "vllm"
-            latest-image: 'latest-gpu-nvidia-cuda-11-vllm'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-transformers'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "transformers"
-            latest-image: 'latest-gpu-nvidia-cuda-11-transformers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-diffusers'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "diffusers"
-            latest-image: 'latest-gpu-nvidia-cuda-11-diffusers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # CUDA 11 additional backends
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-kokoro'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "kokoro"
-            latest-image: 'latest-gpu-nvidia-cuda-11-kokoro'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-faster-whisper'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "faster-whisper"
-            latest-image: 'latest-gpu-nvidia-cuda-11-faster-whisper'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-coqui'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "coqui"
-            latest-image: 'latest-gpu-nvidia-cuda-11-coqui'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-bark'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "bark"
-            latest-image: 'latest-gpu-nvidia-cuda-11-bark'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "11"
-            cuda-minor-version: "7"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-11-chatterbox'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "chatterbox"
-            latest-image: 'latest-gpu-nvidia-cuda-11-chatterbox'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # CUDA 12 builds
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-rerankers'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "rerankers"
-            latest-image: 'latest-gpu-nvidia-cuda-12-rerankers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-vllm'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "vllm"
-            latest-image: 'latest-gpu-nvidia-cuda-12-vllm'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-transformers'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "transformers"
-            latest-image: 'latest-gpu-nvidia-cuda-12-transformers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-diffusers'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "diffusers"
-            latest-image: 'latest-gpu-nvidia-cuda-12-diffusers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # CUDA 12 additional backends
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-kokoro'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "kokoro"
-            latest-image: 'latest-gpu-nvidia-cuda-12-kokoro'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-faster-whisper'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "faster-whisper"
-            latest-image: 'latest-gpu-nvidia-cuda-12-faster-whisper'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-coqui'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "coqui"
-            latest-image: 'latest-gpu-nvidia-cuda-12-coqui'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-bark'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "bark"
-            latest-image: 'latest-gpu-nvidia-cuda-12-bark'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'cublas'
-            cuda-major-version: "12"
-            cuda-minor-version: "0"
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-nvidia-cuda-12-chatterbox'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "chatterbox"
-            latest-image: 'latest-gpu-nvidia-cuda-12-chatterbox'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # hipblas builds
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-rerankers'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "rerankers"
-            latest-image: 'latest-gpu-rocm-hipblas-rerankers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-vllm'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "vllm"
-            latest-image: 'latest-gpu-rocm-hipblas-vllm'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-transformers'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "transformers"
-            latest-image: 'latest-gpu-rocm-hipblas-transformers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-diffusers'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "diffusers"
-            latest-image: 'latest-gpu-rocm-hipblas-diffusers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # ROCm additional backends
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-kokoro'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "kokoro"
-            latest-image: 'latest-gpu-rocm-hipblas-kokoro'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-faster-whisper'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "faster-whisper"
-            latest-image: 'latest-gpu-rocm-hipblas-faster-whisper'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-coqui'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "coqui"
-            latest-image: 'latest-gpu-rocm-hipblas-coqui'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'hipblas'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-rocm-hipblas-bark'
-            runs-on: 'ubuntu-latest'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            backend: "bark"
-            latest-image: 'latest-gpu-rocm-hipblas-bark'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # sycl builds
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-rerankers'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "rerankers"
-            latest-image: 'latest-gpu-intel-sycl-f32-rerankers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f16'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f16-rerankers'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "rerankers"
-            latest-image: 'latest-gpu-intel-sycl-f16-rerankers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-vllm'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "vllm"
-            latest-image: 'latest-gpu-intel-sycl-f32-vllm'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f16'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f16-vllm'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "vllm"
-            latest-image: 'latest-gpu-intel-sycl-f16-vllm'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-transformers'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "transformers"
-            latest-image: 'latest-gpu-intel-sycl-f32-transformers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f16'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f16-transformers'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "transformers"
-            latest-image: 'latest-gpu-intel-sycl-f16-transformers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-diffusers'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "diffusers"
-            latest-image: 'latest-gpu-intel-sycl-f32-diffusers'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # SYCL additional backends
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-kokoro'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "kokoro"
-            latest-image: 'latest-gpu-intel-sycl-f32-kokoro'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f16'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f16-kokoro'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "kokoro"
-            latest-image: 'latest-gpu-intel-sycl-f16-kokoro'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-faster-whisper'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "faster-whisper"
-            latest-image: 'latest-gpu-intel-sycl-f32-faster-whisper'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f16'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f16-faster-whisper'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "faster-whisper"
-            latest-image: 'latest-gpu-intel-sycl-f16-faster-whisper'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-coqui'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "coqui"
-            latest-image: 'latest-gpu-intel-sycl-f32-coqui'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f16'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f16-coqui'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "coqui"
-            latest-image: 'latest-gpu-intel-sycl-f16-coqui'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f32'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f32-bark'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "bark"
-            latest-image: 'latest-gpu-intel-sycl-f32-bark'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          - build-type: 'sycl_f16'
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-gpu-intel-sycl-f16-bark'
-            runs-on: 'ubuntu-latest'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            backend: "bark"
-            latest-image: 'latest-gpu-intel-sycl-f16-bark'
-            dockerfile: "./backend/Dockerfile.python"
-            context: "./backend"
-          # bark-cpp
-          - build-type: ''
-            cuda-major-version: ""
-            cuda-minor-version: ""
-            platforms: 'linux/amd64'
-            tag-latest: 'true'
-            tag-suffix: '-bark-cpp'
-            runs-on: 'ubuntu-latest'
-            base-image: "ubuntu:22.04"
-            backend: "bark"
-            latest-image: 'latest-bark-cpp'
-            dockerfile: "./backend/Dockerfile.go"
-            context: "./"

.github/workflows/backend_build.yml

@@ -1,252 +0,0 @@
----
-name: 'build python backend container images (reusable)'
-
-on:
-  workflow_call:
-    inputs:
-      base-image:
-        description: 'Base image'
-        required: true
-        type: string
-      build-type:
-        description: 'Build type'
-        default: ''
-        type: string
-      cuda-major-version:
-        description: 'CUDA major version'
-        default: "12"
-        type: string
-      cuda-minor-version:
-        description: 'CUDA minor version'
-        default: "1"
-        type: string
-      platforms:
-        description: 'Platforms'
-        default: ''
-        type: string
-      tag-latest:
-        description: 'Tag latest'
-        default: ''
-        type: string
-      latest-image:
-        description: 'Tag latest'
-        default: ''
-        type: string
-      tag-suffix:
-        description: 'Tag suffix'
-        default: ''
-        type: string
-      runs-on:
-        description: 'Runs on'
-        required: true
-        default: ''
-        type: string
-      backend:
-        description: 'Backend to build'
-        required: true
-        type: string
-      context:
-        description: 'Build context'
-        required: true
-        type: string
-      dockerfile:
-        description: 'Build Dockerfile'
-        required: true
-        type: string
-    secrets:
-      dockerUsername:
-        required: true
-      dockerPassword:
-        required: true
-      quayUsername:
-        required: true
-      quayPassword:
-        required: true
-
-jobs:
-  reusable_python_backend-build:
-    runs-on: ${{ inputs.runs-on }}
-    steps:
-
-
-      - name: Free Disk Space (Ubuntu)
-        if: inputs.runs-on == 'ubuntu-latest'
-        uses: jlumbroso/free-disk-space@main
-        with:
-          # this might remove tools that are actually needed,
-          # if set to "true" but frees about 6 GB
-          tool-cache: true
-          # all of these default to true, but feel free to set to
-          # "false" if necessary for your workflow
-          android: true
-          dotnet: true
-          haskell: true
-          large-packages: true
-          docker-images: true
-          swap-storage: true
-
-      - name: Force Install GIT latest
-        run: |
-          sudo apt-get update \
-          && sudo apt-get install -y software-properties-common \
-          && sudo apt-get update \
-          && sudo add-apt-repository -y ppa:git-core/ppa \
-          && sudo apt-get update \
-          && sudo apt-get install -y git
-
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - name: Release space from worker
-        if: inputs.runs-on == 'ubuntu-latest'
-        run: |
-          echo "Listing top largest packages"
-          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
-          head -n 30 <<< "${pkgs}"
-          echo
-          df -h
-          echo
-          sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
-          sudo apt-get remove --auto-remove android-sdk-platform-tools snapd || true
-          sudo apt-get purge --auto-remove android-sdk-platform-tools snapd || true
-          sudo rm -rf /usr/local/lib/android
-          sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
-          sudo rm -rf /usr/share/dotnet
-          sudo apt-get remove -y '^mono-.*' || true
-          sudo apt-get remove -y '^ghc-.*' || true
-          sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
-          sudo apt-get remove -y 'php.*' || true
-          sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
-          sudo apt-get remove -y '^google-.*' || true
-          sudo apt-get remove -y azure-cli || true
-          sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
-          sudo apt-get remove -y '^gfortran-.*' || true
-          sudo apt-get remove -y microsoft-edge-stable || true
-          sudo apt-get remove -y firefox || true
-          sudo apt-get remove -y powershell || true
-          sudo apt-get remove -y r-base-core || true
-          sudo apt-get autoremove -y
-          sudo apt-get clean
-          echo
-          echo "Listing top largest packages"
-          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
-          head -n 30 <<< "${pkgs}"
-          echo
-          sudo rm -rfv build || true
-          sudo rm -rf /usr/share/dotnet || true
-          sudo rm -rf /opt/ghc || true
-          sudo rm -rf "/usr/local/share/boost" || true
-          sudo rm -rf "$AGENT_TOOLSDIRECTORY" || true
-          df -h
-
-      - name: Docker meta
-        id: meta
-        if: github.event_name != 'pull_request'
-        uses: docker/metadata-action@v5
-        with:
-          images: |
-            quay.io/go-skynet/local-ai-backends
-            localai/localai-backends
-          tags: |
-            type=ref,event=branch
-            type=semver,pattern={{raw}}
-            type=sha
-          flavor: |
-            latest=${{ inputs.tag-latest }}
-            suffix=${{ inputs.tag-suffix }}
-
-      - name: Docker meta for PR
-        id: meta_pull_request
-        if: github.event_name == 'pull_request'
-        uses: docker/metadata-action@v5
-        with:
-          images: |
-            quay.io/go-skynet/ci-tests
-          tags: |
-            type=ref,event=branch,suffix=${{ github.event.number }}-${{ inputs.backend }}-${{ inputs.build-type }}-${{ inputs.cuda-major-version }}-${{ inputs.cuda-minor-version }}
-            type=semver,pattern={{raw}},suffix=${{ github.event.number }}-${{ inputs.backend }}-${{ inputs.build-type }}-${{ inputs.cuda-major-version }}-${{ inputs.cuda-minor-version }}
-            type=sha,suffix=${{ github.event.number }}-${{ inputs.backend }}-${{ inputs.build-type }}-${{ inputs.cuda-major-version }}-${{ inputs.cuda-minor-version }}
-          flavor: |
-            latest=${{ inputs.tag-latest }}
-            suffix=${{ inputs.tag-suffix }}
-## End testing image
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@master
-        with:
-          platforms: all
-
-      - name: Set up Docker Buildx
-        id: buildx
-        uses: docker/setup-buildx-action@master
-
-      - name: Login to DockerHub
-        if: github.event_name != 'pull_request'
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.dockerUsername }}
-          password: ${{ secrets.dockerPassword }}
-
-      - name: Login to Quay.io
-        # if: github.event_name != 'pull_request'
-        uses: docker/login-action@v3
-        with:
-          registry: quay.io
-          username: ${{ secrets.quayUsername }}
-          password: ${{ secrets.quayPassword }}
-
-      - name: Build and push
-        uses: docker/build-push-action@v6
-        if: github.event_name != 'pull_request'
-        with:
-          builder: ${{ steps.buildx.outputs.name }}
-          build-args: |
-            BUILD_TYPE=${{ inputs.build-type }}
-            CUDA_MAJOR_VERSION=${{ inputs.cuda-major-version }}
-            CUDA_MINOR_VERSION=${{ inputs.cuda-minor-version }}
-            BASE_IMAGE=${{ inputs.base-image }}
-            BACKEND=${{ inputs.backend }}
-          context: ./backend
-          file: ./backend/Dockerfile.python
-          cache-from: type=gha
-          platforms: ${{ inputs.platforms }}
-          push: ${{ github.event_name != 'pull_request' }}
-          tags: ${{ steps.meta.outputs.tags }}
-          labels: ${{ steps.meta.outputs.labels }}
-
-      - name: Build and push (PR)
-        uses: docker/build-push-action@v6
-        if: github.event_name == 'pull_request'
-        with:
-          builder: ${{ steps.buildx.outputs.name }}
-          build-args: |
-            BUILD_TYPE=${{ inputs.build-type }}
-            CUDA_MAJOR_VERSION=${{ inputs.cuda-major-version }}
-            CUDA_MINOR_VERSION=${{ inputs.cuda-minor-version }}
-            BASE_IMAGE=${{ inputs.base-image }}
-            BACKEND=${{ inputs.backend }}
-          context: ./backend
-          file: ./backend/Dockerfile.python
-          cache-from: type=gha
-          platforms: ${{ inputs.platforms }}
-          push: true
-          tags: ${{ steps.meta_pull_request.outputs.tags }}
-          labels: ${{ steps.meta_pull_request.outputs.labels }}
-
-      - name: Cleanup
-        run: |
-          docker builder prune -f
-          docker system prune --force --volumes --all
-
-      - name: Latest tag
-        if: github.event_name != 'pull_request' && inputs.latest-image != '' && github.ref_type == 'tag'
-        run: |
-          docker pull localai/localai-backends:${{ steps.meta.outputs.version }}
-          docker tag localai/localai-backends:${{ steps.meta.outputs.version }} localai/localai-backends:${{ inputs.latest-image }}
-          docker push localai/localai-backends:${{ inputs.latest-image }}
-          docker pull quay.io/go-skynet/local-ai-backends:${{ steps.meta.outputs.version }}
-          docker tag quay.io/go-skynet/local-ai-backends:${{ steps.meta.outputs.version }} quay.io/go-skynet/local-ai-backends:${{ inputs.latest-image }}
-          docker push quay.io/go-skynet/local-ai-backends:${{ inputs.latest-image }}
-
-      - name: job summary
-        run: |
-          echo "Built image: ${{ steps.meta.outputs.labels }}" >> $GITHUB_STEP_SUMMARY 

.github/workflows/bump_deps.yaml

@@ -12,7 +12,7 @@ jobs:
           - repository: "ggml-org/llama.cpp"
             variable: "CPPLLAMA_VERSION"
             branch: "master"
-          - repository: "ggml-org/whisper.cpp"
+          - repository: "ggerganov/whisper.cpp"
             variable: "WHISPER_CPP_VERSION"
             branch: "master"
           - repository: "PABannier/bark.cpp"

.github/workflows/checksum_checker.yaml

@@ -5,7 +5,7 @@ on:
   workflow_dispatch:
 jobs:
   checksum_check:
-    runs-on: ubuntu-latest
+    runs-on: arc-runner-set
     steps:
       - name: Force Install GIT latest
         run: |

.github/workflows/dependabot_auto.yml

@@ -14,7 +14,7 @@ jobs:
     steps:
       - name: Dependabot metadata
         id: metadata
-        uses: dependabot/fetch-metadata@v2.4.0
+        uses: dependabot/fetch-metadata@v2.3.0
         with:
           github-token: "${{ secrets.GITHUB_TOKEN }}"
           skip-commit-verification: true

.github/workflows/deploy-explorer.yaml

@@ -42,7 +42,7 @@ jobs:
             script: |
                 sudo rm -rf local-ai/ || true
       - name: copy file via ssh
-        uses: appleboy/scp-action@v1.0.0
+        uses: appleboy/scp-action@v0.1.7
         with:
             host: ${{ secrets.EXPLORER_SSH_HOST }}
             username: ${{ secrets.EXPLORER_SSH_USERNAME }}

.github/workflows/generate_grpc_cache.yaml

@@ -17,7 +17,7 @@ jobs:
       matrix:
         include:
           - grpc-base-image: ubuntu:22.04
-            runs-on: 'ubuntu-latest'
+            runs-on: 'arc-runner-set'
             platforms: 'linux/amd64,linux/arm64'
     runs-on: ${{matrix.runs-on}}
     steps:

.github/workflows/generate_intel_image.yaml

@@ -15,7 +15,7 @@ jobs:
     strategy:
       matrix:
         include:
-          - base-image: intel/oneapi-basekit:2025.1.0-0-devel-ubuntu22.04
+          - base-image: intel/oneapi-basekit:2025.0.0-0-devel-ubuntu22.04
             runs-on: 'ubuntu-latest'
             platforms: 'linux/amd64'
     runs-on: ${{matrix.runs-on}}

.github/workflows/image-pr.yml

@@ -9,12 +9,13 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  image-build:
+  extras-image-build:
     uses: ./.github/workflows/image_build.yml
     with:
       tag-latest: ${{ matrix.tag-latest }}
       tag-suffix: ${{ matrix.tag-suffix }}
       ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
       build-type: ${{ matrix.build-type }}
       cuda-major-version: ${{ matrix.cuda-major-version }}
       cuda-minor-version: ${{ matrix.cuda-minor-version }}
@@ -32,9 +33,18 @@ jobs:
       # Pushing with all jobs in parallel
       # eats the bandwidth of all the nodes
       max-parallel: ${{ github.event_name != 'pull_request' && 4 || 8 }}
-      fail-fast: false
       matrix:
         include:
+          # This is basically covered by the AIO test
+          # - build-type: ''
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-ffmpeg'
+          #   ffmpeg: 'true'
+          #   image-type: 'extras'
+          #   runs-on: 'arc-runner-set'
+          #   base-image: "ubuntu:22.04"
+          #   makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
             cuda-minor-version: "0"
@@ -42,32 +52,89 @@ jobs:
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-ffmpeg'
             ffmpeg: 'true'
-            runs-on: 'ubuntu-latest'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
             base-image: "ubuntu:22.04"
             makeflags: "--jobs=3 --output-sync=target"
-          - build-type: 'hipblas'
-            platforms: 'linux/amd64'
-            tag-latest: 'false'
-            tag-suffix: '-hipblas'
-            ffmpeg: 'false'
-            base-image: "rocm/dev-ubuntu-22.04:6.1"
-            grpc-base-image: "ubuntu:22.04"
-            runs-on: 'ubuntu-latest'
-            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'
-            runs-on: 'ubuntu-latest'
-            makeflags: "--jobs=3 --output-sync=target"
-          - build-type: 'vulkan'
-            platforms: 'linux/amd64'
-            tag-latest: 'false'
-            tag-suffix: '-vulkan-ffmpeg-core'
-            ffmpeg: 'true'
-            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: "0"
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-cublas-cuda12-ffmpeg-core'
+          #   ffmpeg: 'true'
+          #   image-type: 'core'
+          #   runs-on: 'ubuntu-latest'
+          #   base-image: "ubuntu:22.04"
+          #   makeflags: "--jobs=4 --output-sync=target"
+          # - build-type: 'vulkan'
+          #   platforms: 'linux/amd64'
+          #   tag-latest: 'false'
+          #   tag-suffix: '-vulkan-ffmpeg-core'
+          #   ffmpeg: 'true'
+          #   image-type: 'core'
+          #   runs-on: 'ubuntu-latest'
+          #   base-image: "ubuntu:22.04"
+          #   makeflags: "--jobs=4 --output-sync=target"

.github/workflows/image.yml

@@ -19,6 +19,7 @@ jobs:
       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 }}
@@ -36,20 +37,230 @@ jobs:
       quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
       quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
     strategy:
+      # Pushing with all jobs in parallel
+      # eats the bandwidth of all the nodes
+      max-parallel: 2
       matrix:
         include:
           - build-type: 'hipblas'
             platforms: 'linux/amd64'
-            tag-latest: 'false'
-            tag-suffix: '-hipblas'
+            tag-latest: 'auto'
+            tag-suffix: '-hipblas-ffmpeg'
             ffmpeg: 'true'
+            image-type: 'extras'
+            aio: "-aio-gpu-hipblas"
             base-image: "rocm/dev-ubuntu-22.04:6.1"
             grpc-base-image: "ubuntu:22.04"
-            runs-on: 'ubuntu-latest'
-            makeflags: "--jobs=3 --output-sync=target"
             latest-image: 'latest-gpu-hipblas'
-            aio: "-aio-gpu-hipblas"
             latest-image-aio: 'latest-aio-gpu-hipblas'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'hipblas'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-hipblas'
+            ffmpeg: 'false'
+            image-type: 'extras'
+            base-image: "rocm/dev-ubuntu-22.04:6.1"
+            grpc-base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'hipblas'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-hipblas-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            base-image: "rocm/dev-ubuntu-22.04:6.1"
+            grpc-base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'hipblas'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-hipblas-core'
+            ffmpeg: 'false'
+            image-type: 'core'
+            base-image: "rocm/dev-ubuntu-22.04:6.1"
+            grpc-base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+  self-hosted-jobs:
+    uses: ./.github/workflows/image_build.yml
+    with:
+      tag-latest: ${{ matrix.tag-latest }}
+      tag-suffix: ${{ matrix.tag-suffix }}
+      ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
+      build-type: ${{ matrix.build-type }}
+      cuda-major-version: ${{ matrix.cuda-major-version }}
+      cuda-minor-version: ${{ matrix.cuda-minor-version }}
+      platforms: ${{ matrix.platforms }}
+      runs-on: ${{ matrix.runs-on }}
+      base-image: ${{ matrix.base-image }}
+      grpc-base-image: ${{ matrix.grpc-base-image }}
+      aio: ${{ matrix.aio }}
+      makeflags: ${{ matrix.makeflags }}
+      latest-image: ${{ matrix.latest-image }}
+      latest-image-aio: ${{ matrix.latest-image-aio }}
+    secrets:
+      dockerUsername: ${{ secrets.DOCKERHUB_USERNAME }}
+      dockerPassword: ${{ secrets.DOCKERHUB_PASSWORD }}
+      quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+      quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
+    strategy:
+      # Pushing with all jobs in parallel
+      # eats the bandwidth of all the nodes
+      max-parallel: ${{ github.event_name != 'pull_request' && 5 || 8 }}
+      matrix:
+        include:
+          # Extra images
+          - build-type: ''
+            #platforms: 'linux/amd64,linux/arm64'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: ''
+            ffmpeg: ''
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: ''
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda11'
+            ffmpeg: ''
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda12'
+            ffmpeg: ''
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-cublas-cuda11-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            aio: "-aio-gpu-nvidia-cuda-11"
+            latest-image: 'latest-gpu-nvidia-cuda-11'
+            latest-image-aio: 'latest-aio-gpu-nvidia-cuda-11'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-cublas-cuda12-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            aio: "-aio-gpu-nvidia-cuda-12"
+            latest-image: 'latest-gpu-nvidia-cuda-12'
+            latest-image-aio: 'latest-aio-gpu-nvidia-cuda-12'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: ''
+            #platforms: 'linux/amd64,linux/arm64'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: ''
+            ffmpeg: ''
+            image-type: 'extras'
+            base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f16'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f16-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            aio: "-aio-gpu-intel-f16"
+            latest-image: 'latest-gpu-intel-f16'
+            latest-image-aio: 'latest-aio-gpu-intel-f16'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f32'
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f32-ffmpeg'
+            ffmpeg: 'true'
+            image-type: 'extras'
+            runs-on: 'arc-runner-set'
+            aio: "-aio-gpu-intel-f32"
+            latest-image: 'latest-gpu-intel-f32'
+            latest-image-aio: 'latest-aio-gpu-intel-f32'
+            makeflags: "--jobs=3 --output-sync=target"
+          # Core images
+          - build-type: 'sycl_f16'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f16-core'
+            ffmpeg: 'false'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f32'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f32-core'
+            ffmpeg: 'false'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f16'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f16-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
+          - build-type: 'sycl_f32'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
+            grpc-base-image: "ubuntu:22.04"
+            tag-suffix: '-sycl-f32-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=3 --output-sync=target"
 
   core-image-build:
     uses: ./.github/workflows/image_build.yml
@@ -57,6 +268,7 @@ jobs:
       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 }}
@@ -75,16 +287,17 @@ jobs:
       quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
       quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
     strategy:
-      #max-parallel: ${{ github.event_name != 'pull_request' && 2 || 4 }}
+      max-parallel: ${{ github.event_name != 'pull_request' && 2 || 4 }}
       matrix:
         include:
           - build-type: ''
             platforms: 'linux/amd64,linux/arm64'
             tag-latest: 'auto'
-            tag-suffix: ''
+            tag-suffix: '-ffmpeg-core'
             ffmpeg: 'true'
+            image-type: 'core'
             base-image: "ubuntu:22.04"
-            runs-on: 'ubuntu-latest'
+            runs-on: 'arc-runner-set'
             aio: "-aio-cpu"
             latest-image: 'latest-cpu'
             latest-image-aio: 'latest-aio-cpu'
@@ -95,72 +308,67 @@ jobs:
             cuda-minor-version: "7"
             platforms: 'linux/amd64'
             tag-latest: 'false'
-            tag-suffix: '-cublas-cuda11'
-            ffmpeg: 'true'
-            runs-on: 'ubuntu-latest'
+            tag-suffix: '-cublas-cuda11-core'
+            ffmpeg: ''
+            image-type: 'core'
             base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
             makeflags: "--jobs=4 --output-sync=target"
             skip-drivers: 'false'
-            latest-image: 'latest-gpu-nvidia-cuda-11'
-            aio: "-aio-gpu-nvidia-cuda-11"
-            latest-image-aio: 'latest-aio-gpu-nvidia-cuda-11'
           - build-type: 'cublas'
             cuda-major-version: "12"
             cuda-minor-version: "0"
             platforms: 'linux/amd64'
             tag-latest: 'false'
-            tag-suffix: '-cublas-cuda12'
+            tag-suffix: '-cublas-cuda12-core'
+            ffmpeg: ''
+            image-type: 'core'
+            base-image: "ubuntu:22.04"
+            runs-on: 'arc-runner-set'
+            makeflags: "--jobs=4 --output-sync=target"
+            skip-drivers: 'false'
+          - build-type: 'cublas'
+            cuda-major-version: "11"
+            cuda-minor-version: "7"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda11-ffmpeg-core'
             ffmpeg: 'true'
-            runs-on: 'ubuntu-latest'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=4 --output-sync=target"
+            skip-drivers: 'false'
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-cublas-cuda12-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
             base-image: "ubuntu:22.04"
             skip-drivers: 'false'
             makeflags: "--jobs=4 --output-sync=target"
-            latest-image: 'latest-gpu-nvidia-cuda-12'
-            aio: "-aio-gpu-nvidia-cuda-12"
-            latest-image-aio: 'latest-aio-gpu-nvidia-cuda-12'
           - build-type: 'vulkan'
             platforms: 'linux/amd64'
             tag-latest: 'false'
-            tag-suffix: '-vulkan'
+            tag-suffix: '-vulkan-ffmpeg-core'
+            latest-image: 'latest-vulkan-ffmpeg-core'
             ffmpeg: 'true'
-            runs-on: 'ubuntu-latest'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
             base-image: "ubuntu:22.04"
             skip-drivers: 'false'
             makeflags: "--jobs=4 --output-sync=target"
-            latest-image: 'latest-gpu-vulkan'
-            aio: "-aio-gpu-vulkan"
-            latest-image-aio: 'latest-aio-gpu-vulkan'
-          - 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: 'true'
-            runs-on: 'ubuntu-latest'
-            makeflags: "--jobs=3 --output-sync=target"
-            latest-image: 'latest-gpu-intel-f16'
-            aio: "-aio-gpu-intel-f16"
-            latest-image-aio: 'latest-aio-gpu-intel-f16'
-          - build-type: 'sycl_f32'
-            platforms: 'linux/amd64'
-            tag-latest: 'false'
-            base-image: "quay.io/go-skynet/intel-oneapi-base:latest"
-            grpc-base-image: "ubuntu:22.04"
-            tag-suffix: '-sycl-f32'
-            ffmpeg: 'true'
-            runs-on: 'ubuntu-latest'
-            makeflags: "--jobs=3 --output-sync=target"
-            latest-image: 'latest-gpu-intel-f32'
-            aio: "-aio-gpu-intel-f32"
-            latest-image-aio: 'latest-aio-gpu-intel-f32'
-
   gh-runner:
     uses: ./.github/workflows/image_build.yml
     with:
       tag-latest: ${{ matrix.tag-latest }}
       tag-suffix: ${{ matrix.tag-suffix }}
       ffmpeg: ${{ matrix.ffmpeg }}
+      image-type: ${{ matrix.image-type }}
       build-type: ${{ matrix.build-type }}
       cuda-major-version: ${{ matrix.cuda-major-version }}
       cuda-minor-version: ${{ matrix.cuda-minor-version }}
@@ -186,10 +394,11 @@ jobs:
             cuda-minor-version: "0"
             platforms: 'linux/arm64'
             tag-latest: 'false'
-            tag-suffix: '-nvidia-l4t-arm64'
-            latest-image: 'latest-nvidia-l4t-arm64'
+            tag-suffix: '-nvidia-l4t-arm64-core'
+            latest-image: 'latest-nvidia-l4t-arm64-core'
             ffmpeg: 'true'
+            image-type: 'core'
             base-image: "nvcr.io/nvidia/l4t-jetpack:r36.4.0"
             runs-on: 'ubuntu-24.04-arm'
             makeflags: "--jobs=4 --output-sync=target"
-            skip-drivers: 'true'
+            skip-drivers: 'true'

.github/workflows/image_build.yml

@@ -53,6 +53,10 @@ on:
         description: 'Skip drivers by default'
         default: 'false'
         type: string
+      image-type:
+        description: 'Image type'
+        default: ''
+        type: string
       runs-on:
         description: 'Runs on'
         required: true
@@ -81,22 +85,6 @@ jobs:
   reusable_image-build:
     runs-on: ${{ inputs.runs-on }}
     steps:
-
-      - name: Free Disk Space (Ubuntu)
-        if: inputs.runs-on == 'ubuntu-latest'
-        uses: jlumbroso/free-disk-space@main
-        with:
-          # this might remove tools that are actually needed,
-          # if set to "true" but frees about 6 GB
-          tool-cache: true
-          # all of these default to true, but feel free to set to
-          # "false" if necessary for your workflow
-          android: true
-          dotnet: true
-          haskell: true
-          large-packages: true
-          docker-images: true
-          swap-storage: true
       - name: Force Install GIT latest
         run: |
           sudo apt-get update \
@@ -118,8 +106,8 @@ jobs:
           df -h
           echo
           sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
-          sudo apt-get remove --auto-remove android-sdk-platform-tools snapd || true
-          sudo apt-get purge --auto-remove android-sdk-platform-tools snapd || true
+          sudo apt-get remove --auto-remove android-sdk-platform-tools || true
+          sudo apt-get purge --auto-remove android-sdk-platform-tools || true
           sudo rm -rf /usr/local/lib/android
           sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
           sudo rm -rf /usr/share/dotnet
@@ -171,11 +159,11 @@ jobs:
         uses: docker/metadata-action@v5
         with:
           images: |
-            quay.io/go-skynet/ci-tests
+            ttl.sh/localai-ci-pr-${{ github.event.number }}
           tags: |
-            type=ref,event=branch,suffix=localai${{ github.event.number }}-${{ inputs.build-type }}-${{ inputs.cuda-major-version }}-${{ inputs.cuda-minor-version }}
-            type=semver,pattern={{raw}},suffix=localai${{ github.event.number }}-${{ inputs.build-type }}-${{ inputs.cuda-major-version }}-${{ inputs.cuda-minor-version }}
-            type=sha,suffix=localai${{ github.event.number }}-${{ inputs.build-type }}-${{ inputs.cuda-major-version }}-${{ inputs.cuda-minor-version }}
+            type=ref,event=branch
+            type=semver,pattern={{raw}}
+            type=sha
           flavor: |
             latest=${{ inputs.tag-latest }}
             suffix=${{ inputs.tag-suffix }}
@@ -244,6 +232,7 @@ jobs:
             CUDA_MAJOR_VERSION=${{ inputs.cuda-major-version }}
             CUDA_MINOR_VERSION=${{ inputs.cuda-minor-version }}
             FFMPEG=${{ inputs.ffmpeg }}
+            IMAGE_TYPE=${{ inputs.image-type }}
             BASE_IMAGE=${{ inputs.base-image }}
             GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
             GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
@@ -272,6 +261,7 @@ jobs:
             CUDA_MAJOR_VERSION=${{ inputs.cuda-major-version }}
             CUDA_MINOR_VERSION=${{ inputs.cuda-minor-version }}
             FFMPEG=${{ inputs.ffmpeg }}
+            IMAGE_TYPE=${{ inputs.image-type }}
             BASE_IMAGE=${{ inputs.base-image }}
             GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
             GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
@@ -282,9 +272,13 @@ jobs:
           file: ./Dockerfile
           cache-from: type=gha
           platforms: ${{ inputs.platforms }}
-          #push: true
+          push: true
           tags: ${{ steps.meta_pull_request.outputs.tags }}
           labels: ${{ steps.meta_pull_request.outputs.labels }}
+      - name: Testing image
+        if: github.event_name == 'pull_request'
+        run: |
+          echo "Image is available at ttl.sh/localai-ci-pr-${{ github.event.number }}:${{ steps.meta_pull_request.outputs.version }}" >> $GITHUB_STEP_SUMMARY
 ## End testing image
       - name: Build and push AIO image
         if: inputs.aio != ''

.github/workflows/notify-models.yaml

@@ -8,7 +8,7 @@ jobs:
   notify-discord:
     if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
     env:
-        MODEL_NAME: gemma-3-12b-it
+        MODEL_NAME: hermes-2-theta-llama-3-8b
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v4
@@ -16,9 +16,9 @@ jobs:
         fetch-depth: 0 # needed to checkout all branches for this Action to work
     - uses: mudler/localai-github-action@v1
       with:
-        model: 'gemma-3-12b-it' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
+        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@v2.8.1
+    - uses: GrantBirki/git-diff-action@v2.8.0
       id: git-diff-action
       with:
             json_diff_file_output: diff.json
@@ -79,7 +79,7 @@ jobs:
         args: ${{ steps.summarize.outputs.message }}
     - name: Setup tmate session if fails
       if: ${{ failure() }}
-      uses: mxschmitt/action-tmate@v3.22
+      uses: mxschmitt/action-tmate@v3.19
       with:
         detached: true
         connect-timeout-seconds: 180
@@ -87,7 +87,7 @@ jobs:
   notify-twitter:
     if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
     env:
-        MODEL_NAME: gemma-3-12b-it
+        MODEL_NAME: hermes-2-theta-llama-3-8b
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v4
@@ -99,7 +99,7 @@ jobs:
         docker run -e -ti -d --name local-ai -p 8080:8080 localai/localai:master-ffmpeg-core run --debug $MODEL_NAME
         until [ "`docker inspect -f {{.State.Health.Status}} local-ai`" == "healthy" ]; do echo "Waiting for container to be ready";  docker logs --tail 10 local-ai; sleep 2; done
       # Check the PR diff using the current branch and the base branch of the PR
-    - uses: GrantBirki/git-diff-action@v2.8.1
+    - uses: GrantBirki/git-diff-action@v2.8.0
       id: git-diff-action
       with:
             json_diff_file_output: diff.json
@@ -161,7 +161,7 @@ jobs:
         TWITTER_ACCESS_TOKEN_SECRET: ${{ secrets.TWITTER_ACCESS_TOKEN_SECRET }}
     - name: Setup tmate session if fails
       if: ${{ failure() }}
-      uses: mxschmitt/action-tmate@v3.22
+      uses: mxschmitt/action-tmate@v3.19
       with:
         detached: true
         connect-timeout-seconds: 180

.github/workflows/notify-releases.yaml

@@ -14,7 +14,7 @@ jobs:
     steps:
     - uses: mudler/localai-github-action@v1
       with:
-        model: 'gemma-3-12b-it' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
+        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: |
@@ -60,4 +60,4 @@ jobs:
         DISCORD_AVATAR: "https://avatars.githubusercontent.com/u/139863280?v=4"
       uses: Ilshidur/action-discord@master
       with:
-        args: ${{ steps.summarize.outputs.message }}
+        args: ${{ steps.summarize.outputs.message }}

.github/workflows/release.yaml

@@ -36,7 +36,6 @@ jobs:
           sudo apt-get update
           sudo apt-get install build-essential ffmpeg protobuf-compiler ccache upx-ucl gawk
           sudo apt-get install -qy binutils-aarch64-linux-gnu gcc-aarch64-linux-gnu g++-aarch64-linux-gnu libgmock-dev
-          make install-go-tools
       - 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
@@ -124,70 +123,14 @@ jobs:
             release/*
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.22
+        uses: mxschmitt/action-tmate@v3.19
         with:
           detached: true
           connect-timeout-seconds: 180
           limit-access-to-actor: true
   build-linux:
-    runs-on: ubuntu-latest
+    runs-on: arc-runner-set
     steps:
-      - name: Free Disk Space (Ubuntu)
-        uses: jlumbroso/free-disk-space@main
-        with:
-          # this might remove tools that are actually needed,
-          # if set to "true" but frees about 6 GB
-          tool-cache: true
-          # all of these default to true, but feel free to set to
-          # "false" if necessary for your workflow
-          android: true
-          dotnet: true
-          haskell: true
-          large-packages: true
-          docker-images: true
-          swap-storage: true
-
-      - name: Release space from worker
-        run: |
-          echo "Listing top largest packages"
-          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
-          head -n 30 <<< "${pkgs}"
-          echo
-          df -h
-          echo
-          sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
-          sudo apt-get remove --auto-remove android-sdk-platform-tools snapd || true
-          sudo apt-get purge --auto-remove android-sdk-platform-tools snapd || true
-          sudo rm -rf /usr/local/lib/android
-          sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
-          sudo rm -rf /usr/share/dotnet
-          sudo apt-get remove -y '^mono-.*' || true
-          sudo apt-get remove -y '^ghc-.*' || true
-          sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
-          sudo apt-get remove -y 'php.*' || true
-          sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
-          sudo apt-get remove -y '^google-.*' || true
-          sudo apt-get remove -y azure-cli || true
-          sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
-          sudo apt-get remove -y '^gfortran-.*' || true
-          sudo apt-get remove -y microsoft-edge-stable || true
-          sudo apt-get remove -y firefox || true
-          sudo apt-get remove -y powershell || true
-          sudo apt-get remove -y r-base-core || true
-          sudo apt-get autoremove -y
-          sudo apt-get clean
-          echo
-          echo "Listing top largest packages"
-          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
-          head -n 30 <<< "${pkgs}"
-          echo
-          sudo rm -rfv build || true
-          sudo rm -rf /usr/share/dotnet || true
-          sudo rm -rf /opt/ghc || true
-          sudo rm -rf "/usr/local/share/boost" || true
-          sudo rm -rf "$AGENT_TOOLSDIRECTORY" || true
-          df -h
-
       - name: Force Install GIT latest
         run: |
           sudo apt-get update \
@@ -208,7 +151,6 @@ jobs:
         run: |
           sudo apt-get update
           sudo apt-get install -y wget curl build-essential ffmpeg protobuf-compiler ccache upx-ucl gawk cmake libgmock-dev
-          make install-go-tools
       - 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
@@ -233,12 +175,17 @@ jobs:
             sudo apt-get update
             sudo DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends ca-certificates curl libnuma-dev gnupg
 
-            sudo apt update
-            wget https://repo.radeon.com/amdgpu-install/6.4.1/ubuntu/noble/amdgpu-install_6.4.60401-1_all.deb
-            sudo apt install ./amdgpu-install_6.4.60401-1_all.deb
-            sudo apt update
+            curl -sL https://repo.radeon.com/rocm/rocm.gpg.key | sudo apt-key add -
 
-            sudo amdgpu-install --usecase=rocm
+            printf "deb [arch=amd64] https://repo.radeon.com/rocm/apt/$ROCM_VERSION/ jammy main" | sudo tee /etc/apt/sources.list.d/rocm.list
+
+            printf "deb [arch=amd64] https://repo.radeon.com/amdgpu/$AMDGPU_VERSION/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
+            printf 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
+            sudo apt-get update
+
+            sudo DEBIAN_FRONTEND=noninteractive apt-get install -y \
+                hipblas-dev rocm-dev \
+                rocblas-dev
 
             sudo apt-get clean
             sudo rm -rf /var/lib/apt/lists/*
@@ -285,7 +232,7 @@ jobs:
             release/*
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.22
+        uses: mxschmitt/action-tmate@v3.19
         with:
           detached: true
           connect-timeout-seconds: 180
@@ -306,7 +253,8 @@ jobs:
       - name: Dependencies
         run: |
           brew install protobuf grpc
-          make install-go-tools
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
       - name: Build
         id: build
         run: |
@@ -327,7 +275,7 @@ jobs:
             release/*
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.22
+        uses: mxschmitt/action-tmate@v3.19
         with:
           detached: true
           connect-timeout-seconds: 180
@@ -347,7 +295,8 @@ jobs:
       - name: Dependencies
         run: |
           brew install protobuf grpc libomp llvm
-          make install-go-tools
+          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: |
@@ -368,7 +317,7 @@ jobs:
             release/*
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.22
+        uses: mxschmitt/action-tmate@v3.19
         with:
           detached: true
           connect-timeout-seconds: 180

.github/workflows/secscan.yaml

@@ -18,7 +18,7 @@ jobs:
         if: ${{ github.actor != 'dependabot[bot]' }}
       - name: Run Gosec Security Scanner
         if: ${{ github.actor != 'dependabot[bot]' }}
-        uses: securego/gosec@v2.22.5
+        uses: securego/gosec@v2.22.0
         with:
           # we let the report trigger content trigger a failure using the GitHub Security features.
           args: '-no-fail -fmt sarif -out results.sarif ./...'

.github/workflows/test-extra.yml

@@ -14,28 +14,6 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  # Requires CUDA
-  # tests-chatterbox-tts:
-  #   runs-on: ubuntu-latest
-  #   steps:
-  #     - name: Clone
-  #       uses: actions/checkout@v4
-  #       with:
-  #         submodules: true
-  #     - name: Dependencies
-  #       run: |
-  #         sudo apt-get update
-  #         sudo apt-get install build-essential ffmpeg
-  #         # Install UV
-  #         curl -LsSf https://astral.sh/uv/install.sh | sh
-  #         sudo apt-get install -y ca-certificates cmake curl patch python3-pip
-  #         sudo apt-get install -y libopencv-dev
-  #         pip install --user --no-cache-dir grpcio-tools==1.64.1
-
-  #     - name: Test chatterbox-tts
-  #       run: |
-  #          make --jobs=5 --output-sync=target -C backend/python/chatterbox
-  #          make --jobs=5 --output-sync=target -C backend/python/chatterbox test
   tests-transformers:
     runs-on: ubuntu-latest
     steps:
@@ -100,26 +78,6 @@ jobs:
           make --jobs=5 --output-sync=target -C backend/python/diffusers
           make --jobs=5 --output-sync=target -C backend/python/diffusers test
 
-  #tests-vllm:
-  #  runs-on: ubuntu-latest
-  #  steps:
-  #    - name: Clone
-  #      uses: actions/checkout@v4
-  #      with:
-  #        submodules: true
-  #    - name: Dependencies
-  #      run: |
-  #        sudo apt-get update
-  #        sudo apt-get install -y build-essential ffmpeg
-  #        sudo apt-get install -y ca-certificates cmake curl patch python3-pip
-  #        sudo apt-get install -y libopencv-dev
-  #        # Install UV
-  #        curl -LsSf https://astral.sh/uv/install.sh | sh
-  #        pip install --user --no-cache-dir grpcio-tools==1.64.1
-  #    - name: Test vllm backend
-  #      run: |
-  #        make --jobs=5 --output-sync=target -C backend/python/vllm
-  #        make --jobs=5 --output-sync=target -C backend/python/vllm test
   # tests-transformers-musicgen:
   #   runs-on: ubuntu-latest
   #   steps:

.github/workflows/test.yml

@@ -71,7 +71,7 @@ jobs:
         run: |
           sudo apt-get update
           sudo apt-get install build-essential ccache upx-ucl curl ffmpeg
-          sudo apt-get install -y libgmock-dev clang
+          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 && \
@@ -96,10 +96,9 @@ jobs:
 
           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 github.com/GeertJohan/go.rice/rice@latest
 
           # The python3-grpc-tools package in 22.04 is too old
-          pip install --user grpcio-tools==1.71.0 grpcio==1.71.0
+          pip install --user grpcio-tools
 
           make -C backend/python/transformers
 
@@ -131,7 +130,7 @@ jobs:
           PATH="$PATH:/root/go/bin" GO_TAGS="tts" make --jobs 5 --output-sync=target test
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.22
+        uses: mxschmitt/action-tmate@v3.19
         with:
           detached: true
           connect-timeout-seconds: 180
@@ -184,11 +183,10 @@ jobs:
           rm protoc.zip
           go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
           go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
-          go install github.com/GeertJohan/go.rice/rice@latest
           PATH="$PATH:$HOME/go/bin" make protogen-go
       - name: Build images
         run: |
-          docker build --build-arg FFMPEG=true --build-arg MAKEFLAGS="--jobs=5 --output-sync=target" -t local-ai:tests -f Dockerfile .
+          docker build --build-arg FFMPEG=true --build-arg IMAGE_TYPE=extras --build-arg EXTRA_BACKENDS=rerankers --build-arg MAKEFLAGS="--jobs=5 --output-sync=target" -t local-ai:tests -f Dockerfile .
           BASE_IMAGE=local-ai:tests DOCKER_AIO_IMAGE=local-ai-aio:test make docker-aio
       - name: Test
         run: |
@@ -196,7 +194,7 @@ jobs:
             make run-e2e-aio
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.22
+        uses: mxschmitt/action-tmate@v3.19
         with:
           detached: true
           connect-timeout-seconds: 180
@@ -223,8 +221,7 @@ jobs:
       - name: Dependencies
         run: |
           brew install protobuf grpc make protoc-gen-go protoc-gen-go-grpc libomp llvm
-          pip install --user --no-cache-dir grpcio-tools==1.71.0 grpcio==1.71.0
-          go install github.com/GeertJohan/go.rice/rice@latest
+          pip install --user --no-cache-dir grpcio-tools
       - name: Test
         run: |
           export C_INCLUDE_PATH=/usr/local/include
@@ -235,7 +232,7 @@ jobs:
           BUILD_TYPE="GITHUB_CI_HAS_BROKEN_METAL" CMAKE_ARGS="-DGGML_F16C=OFF -DGGML_AVX512=OFF -DGGML_AVX2=OFF -DGGML_FMA=OFF" make --jobs 4 --output-sync=target test
       - name: Setup tmate session if tests fail
         if: ${{ failure() }}
-        uses: mxschmitt/action-tmate@v3.22
+        uses: mxschmitt/action-tmate@v3.19
         with:
           detached: true
           connect-timeout-seconds: 180

.github/workflows/yaml-check.yml

@@ -8,7 +8,7 @@ jobs:
     steps:
       - name: 'Checkout'
         uses: actions/checkout@master
-      - name: 'Yamllint model gallery'
+      - name: 'Yamllint'
         uses: karancode/yamllint-github-action@master
         with:
           yamllint_file_or_dir: 'gallery'
@@ -16,11 +16,3 @@ jobs:
           yamllint_comment: true
         env:
           GITHUB_ACCESS_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      - name: 'Yamllint Backend gallery'
-        uses: karancode/yamllint-github-action@master
-        with:
-          yamllint_file_or_dir: 'backend'
-          yamllint_strict: false
-          yamllint_comment: true
-        env:
-          GITHUB_ACCESS_TOKEN: ${{ secrets.GITHUB_TOKEN }}

Dockerfile

@@ -1,9 +1,10 @@
+ARG IMAGE_TYPE=extras
 ARG BASE_IMAGE=ubuntu:22.04
 ARG GRPC_BASE_IMAGE=${BASE_IMAGE}
 ARG INTEL_BASE_IMAGE=${BASE_IMAGE}
 
 # The requirements-core target is common to all images.  It should not be placed in requirements-core unless every single build will use it.
-FROM ${BASE_IMAGE} AS requirements
+FROM ${BASE_IMAGE} AS requirements-core
 
 USER root
 
@@ -14,16 +15,16 @@ ARG TARGETARCH
 ARG TARGETVARIANT
 
 ENV DEBIAN_FRONTEND=noninteractive
+ENV EXTERNAL_GRPC_BACKENDS="coqui:/build/backend/python/coqui/run.sh,transformers:/build/backend/python/transformers/run.sh,rerankers:/build/backend/python/rerankers/run.sh,autogptq:/build/backend/python/autogptq/run.sh,bark:/build/backend/python/bark/run.sh,diffusers:/build/backend/python/diffusers/run.sh,faster-whisper:/build/backend/python/faster-whisper/run.sh,kokoro:/build/backend/python/kokoro/run.sh,vllm:/build/backend/python/vllm/run.sh,exllama2:/build/backend/python/exllama2/run.sh"
 
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
         build-essential \
         ccache \
-        ca-certificates espeak-ng \
+        ca-certificates \
         curl libssl-dev \
         git \
-        git-lfs \
-        unzip upx-ucl python3 python-is-python3 && \
+        unzip upx-ucl && \
     apt-get clean && \
     rm -rf /var/lib/apt/lists/*
 
@@ -44,10 +45,9 @@ EOT
 RUN curl -L -s https://go.dev/dl/go${GO_VERSION}.linux-${TARGETARCH}.tar.gz | tar -C /usr/local -xz
 ENV PATH=$PATH:/root/go/bin:/usr/local/go/bin
 
-# Install grpc compilers and rice
+# Install grpc compilers
 RUN go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2 && \
-    go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af && \
-    go install github.com/GeertJohan/go.rice/rice@latest
+    go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
 
 COPY --chmod=644 custom-ca-certs/* /usr/local/share/ca-certificates/
 RUN update-ca-certificates
@@ -74,12 +74,38 @@ RUN apt-get update && \
 
 WORKDIR /build
 
+###################################
+###################################
+
+# The requirements-extras target is for any builds with IMAGE_TYPE=extras. It should not be placed in this target unless every IMAGE_TYPE=extras build will use it
+FROM requirements-core AS requirements-extras
+
+# Install uv as a system package
+RUN curl -LsSf https://astral.sh/uv/install.sh | UV_INSTALL_DIR=/usr/bin sh
+ENV PATH="/root/.cargo/bin:${PATH}"
+
+RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        espeak-ng \
+        espeak \
+        python3-pip \
+        python-is-python3 \
+        python3-dev llvm \
+        python3-venv && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/* && \
+    pip install --upgrade pip
+
+# Install grpcio-tools (the version in 22.04 is too old)
+RUN pip install --user grpcio-tools
 
 ###################################
 ###################################
 
 # The requirements-drivers target is for BUILD_TYPE specific items.  If you need to install something specific to CUDA, or specific to ROCM, it goes here.
-FROM requirements AS requirements-drivers
+# This target will be built on top of requirements-core or requirements-extras as retermined by the IMAGE_TYPE build-arg
+FROM requirements-${IMAGE_TYPE} AS requirements-drivers
 
 ARG BUILD_TYPE
 ARG CUDA_MAJOR_VERSION=12
@@ -257,45 +283,26 @@ EOT
 ###################################
 ###################################
 
-# Compile backends first in a separate stage
-FROM builder-base AS builder-backends
-
-COPY --from=grpc /opt/grpc /usr/local
-
-WORKDIR /build
-
-COPY ./Makefile .
-COPY ./backend ./backend
-COPY ./go.mod .
-COPY ./go.sum .
-COPY ./.git ./.git
-
-# Some of the Go backends use libs from the main src, we could further optimize the caching by building the CPP backends before here
-COPY ./pkg/grpc ./pkg/grpc
-COPY ./pkg/utils ./pkg/utils
-COPY ./pkg/langchain ./pkg/langchain
-
-RUN ls -l ./
-RUN make backend-assets
-RUN make prepare
-RUN if [ "${TARGETARCH}" = "arm64" ] || [ "${BUILD_TYPE}" = "hipblas" ]; then \
-        SKIP_GRPC_BACKEND="backend-assets/grpc/llama-cpp-avx512 backend-assets/grpc/llama-cpp-avx backend-assets/grpc/llama-cpp-avx2" make grpcs; \
-    else \
-        make grpcs; \
-    fi
-
 # The builder target compiles LocalAI. This target is not the target that will be uploaded to the registry.
 # Adjustments to the build process should likely be made here.
-FROM builder-backends AS builder
+FROM builder-base AS builder
 
+# Install the pre-built GRPC
+COPY --from=grpc /opt/grpc /usr/local
+
+# Rebuild with defaults backends
 WORKDIR /build
 
 COPY . .
+COPY .git .
+
+RUN make prepare
 
 ## Build the binary
-## If we're on arm64 AND using cublas/hipblas, skip some of the llama-compat backends to save space
-## Otherwise just run the normal build
-RUN if [ "${TARGETARCH}" = "arm64" ] || [ "${BUILD_TYPE}" = "hipblas" ]; then \
+## If it's CUDA or hipblas, we want to skip some of the llama-compat backends to save space
+## We only leave the most CPU-optimized variant and the fallback for the cublas/hipblas build
+## (both will use CUDA or hipblas for the actual computation)
+RUN if [ "${BUILD_TYPE}" = "cublas" ] || [ "${BUILD_TYPE}" = "hipblas" ]; then \
         SKIP_GRPC_BACKEND="backend-assets/grpc/llama-cpp-avx512 backend-assets/grpc/llama-cpp-avx backend-assets/grpc/llama-cpp-avx2" make build; \
     else \
         make build; \
@@ -348,6 +355,8 @@ FROM requirements-drivers
 ARG FFMPEG
 ARG BUILD_TYPE
 ARG TARGETARCH
+ARG IMAGE_TYPE=extras
+ARG EXTRA_BACKENDS
 ARG MAKEFLAGS
 
 ENV BUILD_TYPE=${BUILD_TYPE}
@@ -380,19 +389,64 @@ COPY . .
 COPY --from=builder /build/sources ./sources/
 COPY --from=grpc /opt/grpc /usr/local
 
+RUN make prepare-sources
+
 # Copy the binary
 COPY --from=builder /build/local-ai ./
 
 # Copy shared libraries for piper
 COPY --from=builder /build/sources/go-piper/piper-phonemize/pi/lib/* /usr/lib/
 
+# Change the shell to bash so we can use [[ tests below
+SHELL ["/bin/bash", "-c"]
+# We try to strike a balance between individual layer size (as that affects total push time) and total image size
+# Splitting the backends into more groups with fewer items results in a larger image, but a smaller size for the largest layer
+# Splitting the backends into fewer groups with more items results in a smaller image, but a larger size for the largest layer
+
+RUN if [[ ( "${IMAGE_TYPE}" == "extras ")]]; then \
+        apt-get -qq -y install espeak-ng \
+    ; fi
+
+RUN if [[ ( "${EXTRA_BACKENDS}" =~ "coqui" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/coqui \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "faster-whisper" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/faster-whisper \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "diffusers" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/diffusers \
+    ; fi
+
+RUN if [[ ( "${EXTRA_BACKENDS}" =~ "kokoro" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/kokoro \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "exllama2" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/exllama2 \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "transformers" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/transformers \
+    ; fi
+
+RUN if [[ ( "${EXTRA_BACKENDS}" =~ "vllm" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/vllm \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "autogptq" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/autogptq \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "bark" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/bark \
+    ; fi && \
+    if [[ ( "${EXTRA_BACKENDS}" =~ "rerankers" || -z "${EXTRA_BACKENDS}" ) && "$IMAGE_TYPE" == "extras" ]]; then \
+        make -C backend/python/rerankers \
+    ; fi
+
 # Make sure the models directory exists
-RUN mkdir -p /build/models /build/backends
+RUN mkdir -p /build/models
 
 # Define the health check command
 HEALTHCHECK --interval=1m --timeout=10m --retries=10 \
   CMD curl -f ${HEALTHCHECK_ENDPOINT} || exit 1
 
-VOLUME /build/models /build/backends
+VOLUME /build/models
 EXPOSE 8080
 ENTRYPOINT [ "/build/entrypoint.sh" ]

Makefile

@@ -6,11 +6,11 @@ BINARY_NAME=local-ai
 DETECT_LIBS?=true
 
 # llama.cpp versions
-CPPLLAMA_VERSION?=8d947136546773f6410756f37fcc5d3e65b8135d
+CPPLLAMA_VERSION?=10f2e81809bbb69ecfe64fc8b4686285f84b0c07
 
 # whisper.cpp version
-WHISPER_REPO?=https://github.com/ggml-org/whisper.cpp
-WHISPER_CPP_VERSION?=ecb8f3c2b4e282d5ef416516bcbfb92821f06bf6
+WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp
+WHISPER_CPP_VERSION?=6266a9f9e56a5b925e9892acf650f3eb1245814d
 
 # go-piper version
 PIPER_REPO?=https://github.com/mudler/go-piper
@@ -21,11 +21,8 @@ BARKCPP_REPO?=https://github.com/PABannier/bark.cpp.git
 BARKCPP_VERSION?=v1.0.0
 
 # stablediffusion.cpp (ggml)
-STABLEDIFFUSION_GGML_REPO?=https://github.com/richiejp/stable-diffusion.cpp
-STABLEDIFFUSION_GGML_VERSION?=53e3b17eb3d0b5760ced06a1f98320b68b34aaae
-
-# ONEAPI variables for SYCL
-export ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh
+STABLEDIFFUSION_GGML_REPO?=https://github.com/leejet/stable-diffusion.cpp
+STABLEDIFFUSION_GGML_VERSION?=19d876ee300a055629926ff836489901f734f2b7
 
 ONNX_VERSION?=1.20.0
 ONNX_ARCH?=x64
@@ -33,12 +30,8 @@ ONNX_OS?=linux
 
 export BUILD_TYPE?=
 export STABLE_BUILD_TYPE?=$(BUILD_TYPE)
-export CMAKE_ARGS?=-DBUILD_SHARED_LIBS=OFF
-export WHISPER_CMAKE_ARGS?=-DBUILD_SHARED_LIBS=OFF
+export CMAKE_ARGS?=
 export BACKEND_LIBS?=
-export WHISPER_DIR=$(abspath ./sources/whisper.cpp)
-export WHISPER_INCLUDE_PATH=$(WHISPER_DIR)/include:$(WHISPER_DIR)/ggml/include
-export WHISPER_LIBRARY_PATH=$(WHISPER_DIR)/build/src/:$(WHISPER_DIR)/build/ggml/src
 
 CGO_LDFLAGS?=
 CGO_LDFLAGS_WHISPER?=
@@ -88,7 +81,6 @@ endif
 # IF native is false, we add -DGGML_NATIVE=OFF to CMAKE_ARGS
 ifeq ($(NATIVE),false)
 	CMAKE_ARGS+=-DGGML_NATIVE=OFF
-	WHISPER_CMAKE_ARGS+=-DGGML_NATIVE=OFF
 endif
 
 # Detect if we are running on arm64
@@ -116,31 +108,13 @@ ifeq ($(OS),Darwin)
 	# disable metal if on Darwin and any other value is explicitly passed.
 	else ifneq ($(BUILD_TYPE),metal)
 		CMAKE_ARGS+=-DGGML_METAL=OFF
-		WHISPER_CMAKE_ARGS+=-DGGML_METAL=OFF
 		export GGML_NO_ACCELERATE=1
 		export GGML_NO_METAL=1
-		GO_LDFLAGS_WHISPER+=-lggml-blas
-		export WHISPER_LIBRARY_PATH:=$(WHISPER_LIBRARY_PATH):$(WHISPER_DIR)/build/ggml/src/ggml-blas
 	endif
 
 	ifeq ($(BUILD_TYPE),metal)
+#			-lcblas 	removed: it seems to always be listed as a duplicate flag.
 		CGO_LDFLAGS += -framework Accelerate
-		CGO_LDFLAGS_WHISPER+=-lggml-metal -lggml-blas
-		CMAKE_ARGS+=-DGGML_METAL=ON
-		CMAKE_ARGS+=-DGGML_METAL_USE_BF16=ON
-		CMAKE_ARGS+=-DGGML_METAL_EMBED_LIBRARY=ON
-		CMAKE_ARGS+=-DGGML_OPENMP=OFF
-		WHISPER_CMAKE_ARGS+=-DGGML_METAL=ON
-		WHISPER_CMAKE_ARGS+=-DGGML_METAL_USE_BF16=ON
-		WHISPER_CMAKE_ARGS+=-DGGML_METAL_EMBED_LIBRARY=ON
-		WHISPER_CMAKE_ARGS+=-DWHISPER_BUILD_EXAMPLES=OFF
-		WHISPER_CMAKE_ARGS+=-DWHISPER_BUILD_TESTS=OFF
-		WHISPER_CMAKE_ARGS+=-DWHISPER_BUILD_SERVER=OFF
-		WHISPER_CMAKE_ARGS+=-DGGML_OPENMP=OFF
-		export WHISPER_LIBRARY_PATH:=$(WHISPER_LIBRARY_PATH):$(WHISPER_DIR)/build/ggml/src/ggml-metal/:$(WHISPER_DIR)/build/ggml/src/ggml-blas
-	else
-		CGO_LDFLAGS_WHISPER+=-lggml-blas
-		export WHISPER_LIBRARY_PATH:=$(WHISPER_LIBRARY_PATH):$(WHISPER_DIR)/build/ggml/src/ggml-blas
 	endif
 else
 CGO_LDFLAGS_WHISPER+=-lgomp
@@ -152,29 +126,21 @@ ifeq ($(BUILD_TYPE),openblas)
 endif
 
 ifeq ($(BUILD_TYPE),cublas)
-	CGO_LDFLAGS+=-lcublas -lcudart -L$(CUDA_LIBPATH) -L$(CUDA_LIBPATH)/stubs/ -lcuda
+	CGO_LDFLAGS+=-lcublas -lcudart -L$(CUDA_LIBPATH)
 	export GGML_CUDA=1
-	CMAKE_ARGS+=-DGGML_CUDA=ON
-	WHISPER_CMAKE_ARGS+=-DGGML_CUDA=ON
-	CGO_LDFLAGS_WHISPER+=-lcufft -lggml-cuda
-	export WHISPER_LIBRARY_PATH:=$(WHISPER_LIBRARY_PATH):$(WHISPER_DIR)/build/ggml/src/ggml-cuda/
+	CGO_LDFLAGS_WHISPER+=-L$(CUDA_LIBPATH)/stubs/ -lcuda -lcufft
 endif
 
 ifeq ($(BUILD_TYPE),vulkan)
 	CMAKE_ARGS+=-DGGML_VULKAN=1
-	WHISPER_CMAKE_ARGS+=-DGGML_VULKAN=1
-	CGO_LDFLAGS_WHISPER+=-lggml-vulkan -lvulkan
-	export WHISPER_LIBRARY_PATH:=$(WHISPER_LIBRARY_PATH):$(WHISPER_DIR)/build/ggml/src/ggml-vulkan/
 endif
 
 ifneq (,$(findstring sycl,$(BUILD_TYPE)))
 	export GGML_SYCL=1
-	CMAKE_ARGS+=-DGGML_SYCL=ON
 endif
 
 ifeq ($(BUILD_TYPE),sycl_f16)
 	export GGML_SYCL_F16=1
-	CMAKE_ARGS+=-DGGML_SYCL_F16=ON
 endif
 
 ifeq ($(BUILD_TYPE),hipblas)
@@ -185,7 +151,7 @@ ifeq ($(BUILD_TYPE),hipblas)
 	export CC=$(ROCM_HOME)/llvm/bin/clang
 	export STABLE_BUILD_TYPE=
 	export GGML_HIP=1
-	GPU_TARGETS ?= gfx803,gfx900,gfx906,gfx908,gfx90a,gfx942,gfx1010,gfx1030,gfx1032,gfx1100,gfx1101,gfx1102
+	GPU_TARGETS ?= gfx900,gfx906,gfx908,gfx940,gfx941,gfx942,gfx90a,gfx1030,gfx1031,gfx1100,gfx1101
 	AMDGPU_TARGETS ?= "$(GPU_TARGETS)"
 	CMAKE_ARGS+=-DGGML_HIP=ON -DAMDGPU_TARGETS="$(AMDGPU_TARGETS)" -DGPU_TARGETS="$(GPU_TARGETS)"
 	CGO_LDFLAGS += -O3 --rtlib=compiler-rt -unwindlib=libgcc -lhipblas -lrocblas --hip-link -L${ROCM_HOME}/lib/llvm/lib
@@ -225,6 +191,7 @@ ALL_GRPC_BACKENDS+=backend-assets/grpc/whisper
 
 ifeq ($(ONNX_OS),linux)
 ifeq ($(ONNX_ARCH),x64)
+	ALL_GRPC_BACKENDS+=backend-assets/grpc/bark-cpp
 	ALL_GRPC_BACKENDS+=backend-assets/grpc/stablediffusion-ggml
 endif
 endif
@@ -293,7 +260,11 @@ backend/go/image/stablediffusion-ggml/libsd.a: sources/stablediffusion-ggml.cpp
 	$(MAKE) -C backend/go/image/stablediffusion-ggml libsd.a
 
 backend-assets/grpc/stablediffusion-ggml: backend/go/image/stablediffusion-ggml/libsd.a backend-assets/grpc
-	$(MAKE) -C backend/go/image/stablediffusion-ggml CGO_LDFLAGS="$(CGO_LDFLAGS)" stablediffusion-ggml
+	CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/backend/go/image/stablediffusion-ggml/ LIBRARY_PATH=$(CURDIR)/backend/go/image/stablediffusion-ggml/ \
+	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/stablediffusion-ggml ./backend/go/image/stablediffusion-ggml/
+ifneq ($(UPX),)
+	$(UPX) backend-assets/grpc/stablediffusion-ggml
+endif
 
 sources/onnxruntime:
 	mkdir -p sources/onnxruntime
@@ -319,9 +290,8 @@ sources/whisper.cpp:
 	git checkout $(WHISPER_CPP_VERSION) && \
 	git submodule update --init --recursive --depth 1 --single-branch
 
-sources/whisper.cpp/build/src/libwhisper.a: sources/whisper.cpp
-	cd sources/whisper.cpp && cmake $(WHISPER_CMAKE_ARGS) . -B ./build
-	cd sources/whisper.cpp/build && cmake --build . --config Release
+sources/whisper.cpp/libwhisper.a: sources/whisper.cpp
+	cd sources/whisper.cpp && $(MAKE) libwhisper.a libggml.a
 
 get-sources: sources/go-piper sources/stablediffusion-ggml.cpp sources/bark.cpp sources/whisper.cpp backend/cpp/llama/llama.cpp
 
@@ -371,14 +341,8 @@ clean-tests:
 clean-dc: clean
 	cp -r /build/backend-assets /workspace/backend-assets
 
-## Install Go tools
-install-go-tools:
-	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 github.com/GeertJohan/go.rice/rice@latest
-
 ## Build:
-build: prepare backend-assets grpcs install-go-tools ## Build the project
+build: prepare backend-assets grpcs ## Build the project
 	$(info ${GREEN}I local-ai build info:${RESET})
 	$(info ${GREEN}I BUILD_TYPE: ${YELLOW}$(BUILD_TYPE)${RESET})
 	$(info ${GREEN}I GO_TAGS: ${YELLOW}$(GO_TAGS)${RESET})
@@ -388,9 +352,7 @@ ifneq ($(BACKEND_LIBS),)
 	$(MAKE) backend-assets/lib
 	cp -f $(BACKEND_LIBS) backend-assets/lib/
 endif
-	rm -rf $(BINARY_NAME) || true
 	CGO_LDFLAGS="$(CGO_LDFLAGS)" $(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o $(BINARY_NAME) ./
-	rice append --exec $(BINARY_NAME)
 
 build-minimal:
 	BUILD_GRPC_FOR_BACKEND_LLAMA=true GRPC_BACKENDS="backend-assets/grpc/llama-cpp-avx2" GO_TAGS=p2p $(MAKE) build
@@ -452,7 +414,7 @@ run: prepare ## run local-ai
 test-models/testmodel.ggml:
 	mkdir test-models
 	mkdir test-dir
-	wget -q https://huggingface.co/mradermacher/gpt2-alpaca-gpt4-GGUF/resolve/main/gpt2-alpaca-gpt4.Q4_K_M.gguf -O test-models/testmodel.ggml
+	wget -q https://huggingface.co/RichardErkhov/Qwen_-_Qwen2-1.5B-Instruct-gguf/resolve/main/Qwen2-1.5B-Instruct.Q2_K.gguf -O test-models/testmodel.ggml
 	wget -q https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin -O test-models/whisper-en
 	wget -q https://huggingface.co/mudler/all-MiniLM-L6-v2/resolve/main/ggml-model-q4_0.bin -O test-models/bert
 	wget -q https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav -O test-dir/audio.wav
@@ -462,7 +424,6 @@ prepare-test: grpcs
 	cp -rf backend-assets core/http
 	cp tests/models_fixtures/* test-models
 
-## Test targets
 test: prepare test-models/testmodel.ggml grpcs
 	@echo 'Running tests'
 	export GO_TAGS="tts debug"
@@ -537,7 +498,7 @@ protogen: protogen-go protogen-python
 protogen-clean: protogen-go-clean protogen-python-clean
 
 .PHONY: protogen-go
-protogen-go: install-go-tools
+protogen-go:
 	mkdir -p pkg/grpc/proto
 	protoc --experimental_allow_proto3_optional -Ibackend/ --go_out=pkg/grpc/proto/ --go_opt=paths=source_relative --go-grpc_out=pkg/grpc/proto/ --go-grpc_opt=paths=source_relative \
     backend/backend.proto
@@ -548,10 +509,18 @@ protogen-go-clean:
 	$(RM) bin/*
 
 .PHONY: protogen-python
-protogen-python: bark-protogen coqui-protogen chatterbox-protogen diffusers-protogen exllama2-protogen rerankers-protogen transformers-protogen kokoro-protogen vllm-protogen faster-whisper-protogen
+protogen-python: autogptq-protogen bark-protogen coqui-protogen diffusers-protogen exllama2-protogen rerankers-protogen transformers-protogen kokoro-protogen vllm-protogen faster-whisper-protogen
 
 .PHONY: protogen-python-clean
-protogen-python-clean: bark-protogen-clean coqui-protogen-clean chatterbox-protogen-clean diffusers-protogen-clean  exllama2-protogen-clean rerankers-protogen-clean transformers-protogen-clean kokoro-protogen-clean vllm-protogen-clean faster-whisper-protogen-clean
+protogen-python-clean: autogptq-protogen-clean bark-protogen-clean coqui-protogen-clean diffusers-protogen-clean  exllama2-protogen-clean rerankers-protogen-clean transformers-protogen-clean kokoro-protogen-clean vllm-protogen-clean faster-whisper-protogen-clean
+
+.PHONY: autogptq-protogen
+autogptq-protogen:
+	$(MAKE) -C backend/python/autogptq protogen
+
+.PHONY: autogptq-protogen-clean
+autogptq-protogen-clean:
+	$(MAKE) -C backend/python/autogptq protogen-clean
 
 .PHONY: bark-protogen
 bark-protogen:
@@ -573,18 +542,10 @@ coqui-protogen-clean:
 diffusers-protogen:
 	$(MAKE) -C backend/python/diffusers protogen
 
-.PHONY: chatterbox-protogen
-chatterbox-protogen:
-	$(MAKE) -C backend/python/chatterbox protogen
-
 .PHONY: diffusers-protogen-clean
 diffusers-protogen-clean:
 	$(MAKE) -C backend/python/diffusers protogen-clean
 
-.PHONY: chatterbox-protogen-clean
-chatterbox-protogen-clean:
-	$(MAKE) -C backend/python/chatterbox protogen-clean
-
 .PHONY: faster-whisper-protogen
 faster-whisper-protogen:
 	$(MAKE) -C backend/python/faster-whisper protogen
@@ -636,10 +597,10 @@ vllm-protogen-clean:
 ## GRPC
 # Note: it is duplicated in the Dockerfile
 prepare-extra-conda-environments: protogen-python
+	$(MAKE) -C backend/python/autogptq
 	$(MAKE) -C backend/python/bark
 	$(MAKE) -C backend/python/coqui
 	$(MAKE) -C backend/python/diffusers
-	$(MAKE) -C backend/python/chatterbox
 	$(MAKE) -C backend/python/faster-whisper
 	$(MAKE) -C backend/python/vllm
 	$(MAKE) -C backend/python/rerankers
@@ -650,14 +611,10 @@ prepare-extra-conda-environments: protogen-python
 prepare-test-extra: protogen-python
 	$(MAKE) -C backend/python/transformers
 	$(MAKE) -C backend/python/diffusers
-	$(MAKE) -C backend/python/chatterbox
-	$(MAKE) -C backend/python/vllm
 
 test-extra: prepare-test-extra
 	$(MAKE) -C backend/python/transformers test
 	$(MAKE) -C backend/python/diffusers test
-	$(MAKE) -C backend/python/chatterbox test
-	$(MAKE) -C backend/python/vllm test
 
 backend-assets:
 	mkdir -p backend-assets
@@ -799,8 +756,8 @@ ifneq ($(UPX),)
 	$(UPX) backend-assets/grpc/silero-vad
 endif
 
-backend-assets/grpc/whisper: sources/whisper.cpp sources/whisper.cpp/build/src/libwhisper.a backend-assets/grpc
-	CGO_LDFLAGS="$(CGO_LDFLAGS) $(CGO_LDFLAGS_WHISPER)" C_INCLUDE_PATH="${WHISPER_INCLUDE_PATH}" LIBRARY_PATH="${WHISPER_LIBRARY_PATH}" LD_LIBRARY_PATH="${WHISPER_LIBRARY_PATH}" \
+backend-assets/grpc/whisper: sources/whisper.cpp sources/whisper.cpp/libwhisper.a backend-assets/grpc
+	CGO_LDFLAGS="$(CGO_LDFLAGS) $(CGO_LDFLAGS_WHISPER)" C_INCLUDE_PATH="$(CURDIR)/sources/whisper.cpp/include:$(CURDIR)/sources/whisper.cpp/ggml/include" LIBRARY_PATH=$(CURDIR)/sources/whisper.cpp \
 	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/whisper ./backend/go/transcribe/whisper
 ifneq ($(UPX),)
 	$(UPX) backend-assets/grpc/whisper
@@ -852,17 +809,17 @@ docker-aio-all:
 
 docker-image-intel:
 	docker build \
-		--build-arg BASE_IMAGE=intel/oneapi-basekit:2025.1.0-0-devel-ubuntu24.04 \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2025.0.0-0-devel-ubuntu22.04 \
 		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
-		--build-arg GO_TAGS="$(GO_TAGS)" \
+		--build-arg GO_TAGS="none" \
 		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
 		--build-arg BUILD_TYPE=sycl_f32 -t $(DOCKER_IMAGE) .
 
 docker-image-intel-xpu:
 	docker build \
-		--build-arg BASE_IMAGE=intel/oneapi-basekit:2025.1.0-0-devel-ubuntu22.04 \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2025.0.0-0-devel-ubuntu22.04 \
 		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
-		--build-arg GO_TAGS="$(GO_TAGS)" \
+		--build-arg GO_TAGS="none" \
 		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
 		--build-arg BUILD_TYPE=sycl_f32 -t $(DOCKER_IMAGE) .
 

README.md

@@ -1,6 +1,7 @@
 <h1 align="center">
   <br>
-  <img height="300" src="./core/http/static/logo.png"> <br>
+  <img height="300" src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd"> <br>
+    LocalAI
 <br>
 </h1>
 
@@ -30,7 +31,7 @@
 
 <p align="center">
 <a href="https://twitter.com/LocalAI_API" target="blank">
-<img src="https://img.shields.io/badge/X-%23000000.svg?style=for-the-badge&logo=X&logoColor=white&label=LocalAI_API" alt="Follow LocalAI_API"/>
+<img src="https://img.shields.io/twitter/follow/LocalAI_API?label=Follow: LocalAI_API&style=social" alt="Follow LocalAI_API"/>
 </a>
 <a href="https://discord.gg/uJAeKSAGDy" target="blank">
 <img src="https://dcbadge.vercel.app/api/server/uJAeKSAGDy?style=flat-square&theme=default-inverted" alt="Join LocalAI Discord Community"/>
@@ -43,154 +44,35 @@
 
 > :bulb: Get help - [❓FAQ](https://localai.io/faq/) [💭Discussions](https://github.com/go-skynet/LocalAI/discussions) [:speech_balloon: Discord](https://discord.gg/uJAeKSAGDy) [:book: Documentation website](https://localai.io/)
 >
-> [💻 Quickstart](https://localai.io/basics/getting_started/) [🖼️ Models](https://models.localai.io/) [🚀 Roadmap](https://github.com/mudler/LocalAI/issues?q=is%3Aissue+is%3Aopen+label%3Aroadmap) [🥽 Demo](https://demo.localai.io) [🌍 Explorer](https://explorer.localai.io) [🛫 Examples](https://github.com/mudler/LocalAI-examples) Try on 
-[![Telegram](https://img.shields.io/badge/Telegram-2CA5E0?style=for-the-badge&logo=telegram&logoColor=white)](https://t.me/localaiofficial_bot)
+> [💻 Quickstart](https://localai.io/basics/getting_started/) [🖼️ Models](https://models.localai.io/) [🚀 Roadmap](https://github.com/mudler/LocalAI/issues?q=is%3Aissue+is%3Aopen+label%3Aroadmap) [🥽 Demo](https://demo.localai.io) [🌍 Explorer](https://explorer.localai.io) [🛫 Examples](https://github.com/mudler/LocalAI-examples) 
 
 [![tests](https://github.com/go-skynet/LocalAI/actions/workflows/test.yml/badge.svg)](https://github.com/go-skynet/LocalAI/actions/workflows/test.yml)[![Build and Release](https://github.com/go-skynet/LocalAI/actions/workflows/release.yaml/badge.svg)](https://github.com/go-skynet/LocalAI/actions/workflows/release.yaml)[![build container images](https://github.com/go-skynet/LocalAI/actions/workflows/image.yml/badge.svg)](https://github.com/go-skynet/LocalAI/actions/workflows/image.yml)[![Bump dependencies](https://github.com/go-skynet/LocalAI/actions/workflows/bump_deps.yaml/badge.svg)](https://github.com/go-skynet/LocalAI/actions/workflows/bump_deps.yaml)[![Artifact Hub](https://img.shields.io/endpoint?url=https://artifacthub.io/badge/repository/localai)](https://artifacthub.io/packages/search?repo=localai)
 
-**LocalAI** is the free, Open Source OpenAI alternative. LocalAI act as a drop-in replacement REST API that's compatible with OpenAI (Elevenlabs, Anthropic... ) API specifications for local AI inferencing. It allows you to run LLMs, generate images, audio (and not only) locally or on-prem with consumer grade hardware, supporting multiple model families. Does not require GPU. It is created and maintained by [Ettore Di Giacinto](https://github.com/mudler).
+**LocalAI** is the free, Open Source OpenAI alternative. LocalAI act as a drop-in replacement REST API that’s compatible with OpenAI (Elevenlabs, Anthropic... ) API specifications for local AI inferencing. It allows you to run LLMs, generate images, audio (and not only) locally or on-prem with consumer grade hardware, supporting multiple model families. Does not require GPU. It is created and maintained by [Ettore Di Giacinto](https://github.com/mudler).
 
-
-## 📚🆕 Local Stack Family
-
-🆕 LocalAI is now part of a comprehensive suite of AI tools designed to work together:
-
-<table>
-  <tr>
-    <td width="50%" valign="top">
-      <a href="https://github.com/mudler/LocalAGI">
-        <img src="https://raw.githubusercontent.com/mudler/LocalAGI/refs/heads/main/webui/react-ui/public/logo_2.png" width="300" alt="LocalAGI Logo">
-      </a>
-    </td>
-    <td width="50%" valign="top">
-      <h3><a href="https://github.com/mudler/LocalAGI">LocalAGI</a></h3>
-      <p>A powerful Local AI agent management platform that serves as a drop-in replacement for OpenAI's Responses API, enhanced with advanced agentic capabilities.</p>
-    </td>
-  </tr>
-  <tr>
-    <td width="50%" valign="top">
-      <a href="https://github.com/mudler/LocalRecall">
-        <img src="https://raw.githubusercontent.com/mudler/LocalRecall/refs/heads/main/static/localrecall_horizontal.png" width="300" alt="LocalRecall Logo">
-      </a>
-    </td>
-    <td width="50%" valign="top">
-      <h3><a href="https://github.com/mudler/LocalRecall">LocalRecall</a></h3>
-      <p>A REST-ful API and knowledge base management system that provides persistent memory and storage capabilities for AI agents.</p>
-    </td>
-  </tr>
-</table>
-
-## Screenshots
-
-
-| Talk Interface | Generate Audio |
-| --- | --- |
-| ![Screenshot 2025-03-31 at 12-01-36 LocalAI - Talk](./docs/assets/images/screenshots/screenshot_tts.png) | ![Screenshot 2025-03-31 at 12-01-29 LocalAI - Generate audio with voice-en-us-ryan-low](./docs/assets/images/screenshots/screenshot_tts.png) |
-
-| Models Overview | Generate Images |
-| --- | --- |
-| ![Screenshot 2025-03-31 at 12-01-20 LocalAI - Models](./docs/assets/images/screenshots/screenshot_gallery.png) | ![Screenshot 2025-03-31 at 12-31-41 LocalAI - Generate images with flux 1-dev](./docs/assets/images/screenshots/screenshot_image.png) |
-
-| Chat Interface | Home |
-| --- | --- |
-| ![Screenshot 2025-03-31 at 11-57-44 LocalAI - Chat with localai-functioncall-qwen2 5-7b-v0 5](./docs/assets/images/screenshots/screenshot_chat.png) | ![Screenshot 2025-03-31 at 11-57-23 LocalAI API - c2a39e3 (c2a39e3639227cfd94ffffe9f5691239acc275a8)](./docs/assets/images/screenshots/screenshot_home.png) |
-
-| Login | Swarm |
-| --- | --- |
-|![Screenshot 2025-03-31 at 12-09-59 ](./docs/assets/images/screenshots/screenshot_login.png) | ![Screenshot 2025-03-31 at 12-10-39 LocalAI - P2P dashboard](./docs/assets/images/screenshots/screenshot_p2p.png) |
-
-## 💻 Quickstart
+![screen](https://github.com/mudler/LocalAI/assets/2420543/20b5ccd2-8393-44f0-aaf6-87a23806381e)
 
 Run the installer script:
 
 ```bash
-# Basic installation
 curl https://localai.io/install.sh | sh
 ```
 
-For more installation options, see [Installer Options](https://localai.io/docs/advanced/installer/).
-
 Or run with docker:
-
-### CPU only image:
-
 ```bash
-docker run -ti --name local-ai -p 8080:8080 localai/localai:latest
-```
+# CPU only image:
+docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-cpu
 
-### NVIDIA GPU Images:
-
-```bash
-# CUDA 12.0 with core features
+# Nvidia GPU:
 docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-nvidia-cuda-12
 
-# CUDA 12.0 with extra Python dependencies
-docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-nvidia-cuda-12-extras
+# CPU and GPU image (bigger size):
+docker run -ti --name local-ai -p 8080:8080 localai/localai:latest
 
-# CUDA 11.7 with core features
-docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-nvidia-cuda-11
-
-# CUDA 11.7 with extra Python dependencies
-docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-nvidia-cuda-11-extras
-
-# NVIDIA Jetson (L4T) ARM64
-docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-nvidia-l4t-arm64
-```
-
-### AMD GPU Images (ROCm):
-
-```bash
-# ROCm with core features
-docker run -ti --name local-ai -p 8080:8080 --device=/dev/kfd --device=/dev/dri --group-add=video localai/localai:latest-gpu-hipblas
-
-# ROCm with extra Python dependencies
-docker run -ti --name local-ai -p 8080:8080 --device=/dev/kfd --device=/dev/dri --group-add=video localai/localai:latest-gpu-hipblas-extras
-```
-
-### Intel GPU Images (oneAPI):
-
-```bash
-# Intel GPU with FP16 support
-docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-gpu-intel-f16
-
-# Intel GPU with FP16 support and extra dependencies
-docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-gpu-intel-f16-extras
-
-# Intel GPU with FP32 support
-docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-gpu-intel-f32
-
-# Intel GPU with FP32 support and extra dependencies
-docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-gpu-intel-f32-extras
-```
-
-### Vulkan GPU Images:
-
-```bash
-# Vulkan with core features
-docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-gpu-vulkan
-```
-
-### AIO Images (pre-downloaded models):
-
-```bash
-# CPU version
+# AIO images (it will pre-download a set of models ready for use, see https://localai.io/basics/container/)
 docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-aio-cpu
-
-# NVIDIA CUDA 12 version
-docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-aio-gpu-nvidia-cuda-12
-
-# NVIDIA CUDA 11 version
-docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-aio-gpu-nvidia-cuda-11
-
-# Intel GPU version
-docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-aio-gpu-intel-f16
-
-# AMD GPU version
-docker run -ti --name local-ai -p 8080:8080 --device=/dev/kfd --device=/dev/dri --group-add=video localai/localai:latest-aio-gpu-hipblas
 ```
 
-For more information about the AIO images and pre-downloaded models, see [Container Documentation](https://localai.io/basics/container/).
-
 To load models:
 
 ```bash
@@ -206,17 +88,10 @@ local-ai run https://gist.githubusercontent.com/.../phi-2.yaml
 local-ai run oci://localai/phi-2:latest
 ```
 
-For more information, see [💻 Getting started](https://localai.io/basics/getting_started/index.html)
+[💻 Getting started](https://localai.io/basics/getting_started/index.html)
 
 ## 📰 Latest project news
 
-- June 2025: [Backend management](https://github.com/mudler/LocalAI/pull/5607) has been added. Attention: extras images are going to be deprecated from the next release! Read [the backend management PR](https://github.com/mudler/LocalAI/pull/5607).
-- May 2025: [Audio input](https://github.com/mudler/LocalAI/pull/5466) and [Reranking](https://github.com/mudler/LocalAI/pull/5396) in llama.cpp backend, [Realtime API](https://github.com/mudler/LocalAI/pull/5392),  Support to Gemma, SmollVLM, and more multimodal models (available in the gallery).
-- May 2025: Important: image name changes [See release](https://github.com/mudler/LocalAI/releases/tag/v2.29.0)
-- Apr 2025: Rebrand, WebUI enhancements
-- Apr 2025: [LocalAGI](https://github.com/mudler/LocalAGI) and [LocalRecall](https://github.com/mudler/LocalRecall) join the LocalAI family stack.
-- Apr 2025: WebUI overhaul, AIO images updates
-- Feb 2025: Backend cleanup, Breaking changes, new backends (kokoro, OutelTTS, faster-whisper), Nvidia L4T images
 - Jan 2025: LocalAI model release: https://huggingface.co/mudler/LocalAI-functioncall-phi-4-v0.3, SANA support in diffusers: https://github.com/mudler/LocalAI/pull/4603
 - Dec 2024: stablediffusion.cpp backend (ggml) added ( https://github.com/mudler/LocalAI/pull/4289 )
 - Nov 2024: Bark.cpp backend added ( https://github.com/mudler/LocalAI/pull/4287 )
@@ -230,6 +105,19 @@ For more information, see [💻 Getting started](https://localai.io/basics/getti
 
 Roadmap items: [List of issues](https://github.com/mudler/LocalAI/issues?q=is%3Aissue+is%3Aopen+label%3Aroadmap)
 
+## 🔥🔥 Hot topics (looking for help):
+
+- Multimodal with vLLM and Video understanding: https://github.com/mudler/LocalAI/pull/3729
+- Realtime API https://github.com/mudler/LocalAI/issues/3714
+- WebUI improvements: https://github.com/mudler/LocalAI/issues/2156
+- Backends v2: https://github.com/mudler/LocalAI/issues/1126
+- Improving UX v2: https://github.com/mudler/LocalAI/issues/1373
+- Assistant API: https://github.com/mudler/LocalAI/issues/1273
+- 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
+
 ## 🚀 [Features](https://localai.io/features/)
 
 - 📖 [Text generation with GPTs](https://localai.io/features/text-generation/) (`llama.cpp`, `transformers`, `vllm` ... [:book: and more](https://localai.io/model-compatibility/index.html#model-compatibility-table))
@@ -243,10 +131,12 @@ Roadmap items: [List of issues](https://github.com/mudler/LocalAI/issues?q=is%3A
 - 🥽 [Vision API](https://localai.io/features/gpt-vision/)
 - 📈 [Reranker API](https://localai.io/features/reranker/)
 - 🆕🖧 [P2P Inferencing](https://localai.io/features/distribute/)
-- [Agentic capabilities](https://github.com/mudler/LocalAGI)
 - 🔊 Voice activity detection (Silero-VAD support)
 - 🌍 Integrated WebUI!
 
+## 💻 Usage
+
+Check out the [Getting started](https://localai.io/basics/getting_started/index.html) section in our documentation.
 
 ### 🔗 Community and integrations
 

aio/cpu/embeddings.yaml

@@ -1,7 +1,7 @@
-embeddings: true
 name: text-embedding-ada-002
+embeddings: true
 parameters:
-  model: huggingface://bartowski/granite-embedding-107m-multilingual-GGUF/granite-embedding-107m-multilingual-f16.gguf
+  model: huggingface://hugging-quants/Llama-3.2-1B-Instruct-Q4_K_M-GGUF/llama-3.2-1b-instruct-q4_k_m.gguf
 
 usage: |
     You can test this model with curl like this:

aio/cpu/rerank.yaml

@@ -1,13 +1,7 @@
 name: jina-reranker-v1-base-en
-reranking: true
-f16: true
+backend: rerankers
 parameters:
-  model: jina-reranker-v1-tiny-en.f16.gguf
-
-download_files:
-  - filename: jina-reranker-v1-tiny-en.f16.gguf
-    sha256: 5f696cf0d0f3d347c4a279eee8270e5918554cdac0ed1f632f2619e4e8341407
-    uri: huggingface://mradermacher/jina-reranker-v1-tiny-en-GGUF/jina-reranker-v1-tiny-en.f16.gguf 
+  model: cross-encoder
 
 usage: |
     You can test this model with curl like this:

aio/cpu/text-to-text.yaml

@@ -1,57 +1,101 @@
-context_size: 8192
-f16: true
-function:
-  grammar:
-    no_mixed_free_string: true
-    schema_type: llama3.1 # or JSON is supported too (json)
-  response_regex:
-  - <function=(?P<name>\w+)>(?P<arguments>.*)</function>
-mmap: true
 name: gpt-4
+mmap: true
 parameters:
-  model: Hermes-3-Llama-3.2-3B-Q4_K_M.gguf
+  model: huggingface://NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+context_size: 8192
+
 stopwords:
-- <|im_end|>
-- <dummy32000>
-- <|eot_id|>
-- <|end_of_text|>
+- "<|im_end|>"
+- "<dummy32000>"
+- "</tool_call>"
+- "<|eot_id|>"
+- "<|end_of_text|>"
+
+function:
+  # disable injecting the "answer" tool
+  disable_no_action: true
+
+  grammar:
+    # This allows the grammar to also return messages
+    mixed_mode: true
+    # Suffix to add to the grammar
+    #prefix: '<tool_call>\n'
+    # Force parallel calls in the grammar
+    # parallel_calls: true
+
+  return_name_in_function_response: true
+  # Without grammar uncomment the lines below
+  # Warning: this is relying only on the capability of the
+  # LLM model to generate the correct function call.
+  json_regex_match: 
+   - "(?s)<tool_call>(.*?)</tool_call>"
+   - "(?s)<tool_call>(.*?)"
+  replace_llm_results:
+  # Drop the scratchpad content from responses
+  - key: "(?s)<scratchpad>.*</scratchpad>"
+    value: ""
+  replace_function_results: 
+  # Replace everything that is not JSON array or object
+  # 
+  - key: '(?s)^[^{\[]*'
+    value: ""
+  - key: '(?s)[^}\]]*$'
+    value: ""
+  - key: "'([^']*?)'"
+    value: "_DQUOTE_${1}_DQUOTE_"
+  - key: '\\"'
+    value: "__TEMP_QUOTE__"
+  - key: "\'"
+    value: "'"
+  - key: "_DQUOTE_"
+    value: '"'
+  - key: "__TEMP_QUOTE__"
+    value: '"'
+  # Drop the scratchpad content from responses
+  - key: "(?s)<scratchpad>.*</scratchpad>"
+    value: ""
+
 template:
   chat: |
-    <|begin_of_text|><|start_header_id|>system<|end_header_id|>
-    You are a helpful assistant<|eot_id|><|start_header_id|>user<|end_header_id|>
-    {{.Input }}
-    <|start_header_id|>assistant<|end_header_id|>
+    {{.Input -}}
+    <|im_start|>assistant
   chat_message: |
-    <|start_header_id|>{{if eq .RoleName "assistant"}}assistant{{else if eq .RoleName "system"}}system{{else if eq .RoleName "tool"}}tool{{else if eq .RoleName "user"}}user{{end}}<|end_header_id|>
-    {{ if .FunctionCall -}}
-    {{ else if eq .RoleName "tool" -}}
-    The Function was executed and the response was:
-    {{ end -}}
-    {{ if .Content -}}
-    {{.Content -}}
-    {{ else if .FunctionCall -}}
-    {{ range .FunctionCall }}
-    [{{.FunctionCall.Name}}({{.FunctionCall.Arguments}})]
-    {{ end }}
-    {{ end -}}
-    <|eot_id|>
+    <|im_start|>{{if eq .RoleName "assistant"}}assistant{{else if eq .RoleName "system"}}system{{else if eq .RoleName "tool"}}tool{{else if eq .RoleName "user"}}user{{end}}
+    {{- if .FunctionCall }}
+    <tool_call>
+    {{- else if eq .RoleName "tool" }}
+    <tool_response>
+    {{- end }}
+    {{- if .Content}}
+    {{.Content }}
+    {{- end }}
+    {{- if .FunctionCall}}
+    {{toJson .FunctionCall}}
+    {{- end }}
+    {{- if .FunctionCall }}
+    </tool_call>
+    {{- else if eq .RoleName "tool" }}
+    </tool_response>
+    {{- end }}<|im_end|>
   completion: |
     {{.Input}}
-  function: |
-    <|start_header_id|>system<|end_header_id|>
-    You are an expert in composing functions. You are given a question and a set of possible functions.
-    Based on the question, you will need to make one or more function/tool calls to achieve the purpose.
-    If none of the functions can be used, point it out. If the given question lacks the parameters required by the function, also point it out. You should only return the function call in tools call sections.
-    If you decide to invoke any of the function(s), you MUST put it in the format as follows:
-    [func_name1(params_name1=params_value1,params_name2=params_value2,...),func_name2(params_name1=params_value1,params_name2=params_value2,...)]
-    You SHOULD NOT include any other text in the response.
-    Here is a list of functions in JSON format that you can invoke.
-    {{toJson .Functions}}
-    <|eot_id|><|start_header_id|>user<|end_header_id|>
-    {{.Input}}
-    <|eot_id|><|start_header_id|>assistant<|end_header_id|>
-
-download_files:
-- filename: Hermes-3-Llama-3.2-3B-Q4_K_M.gguf
-  sha256: 2e220a14ba4328fee38cf36c2c068261560f999fadb5725ce5c6d977cb5126b5
-  uri: huggingface://bartowski/Hermes-3-Llama-3.2-3B-GGUF/Hermes-3-Llama-3.2-3B-Q4_K_M.gguf
+  function: |-
+    <|im_start|>system
+    You are a function calling AI model.
+    Here are the available tools:
+    <tools>
+    {{range .Functions}}
+    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
+    {{end}}
+    </tools>
+    You should call the tools provided to you sequentially
+    Please use <scratchpad> XML tags to record your reasoning and planning before you call the functions as follows:
+    <scratchpad>
+    {step-by-step reasoning and plan in bullet points}
+    </scratchpad>
+    For each function call return a json object with function name and arguments within <tool_call> XML tags as follows:
+    <tool_call>
+    {"arguments": <args-dict>, "name": <function-name>}
+    </tool_call><|im_end|>
+    {{.Input -}}
+    <|im_start|>assistant

aio/cpu/vision.yaml

@@ -1,49 +1,31 @@
+backend: llama-cpp
 context_size: 4096
 f16: true
 mmap: true
-mmproj: minicpm-v-2_6-mmproj-f16.gguf
 name: gpt-4o
+
+roles:
+  user: "USER:"
+  assistant: "ASSISTANT:"
+  system: "SYSTEM:"
+
+mmproj: bakllava-mmproj.gguf
 parameters:
-  model: minicpm-v-2_6-Q4_K_M.gguf
-stopwords:
-- <|im_end|>
-- <dummy32000>
-- </s>
-- <|endoftext|>
+  model: bakllava.gguf
+
 template:
   chat: |
-    {{.Input -}}
-    <|im_start|>assistant
-  chat_message: |
-    <|im_start|>{{ .RoleName }}
-    {{ if .FunctionCall -}}
-    Function call:
-    {{ else if eq .RoleName "tool" -}}
-    Function response:
-    {{ end -}}
-    {{ if .Content -}}
-    {{.Content }}
-    {{ end -}}
-    {{ if .FunctionCall -}}
-    {{toJson .FunctionCall}}
-    {{ end -}}<|im_end|>
-  completion: |
+    A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.
     {{.Input}}
-  function: |
-    <|im_start|>system
-    You are a function calling AI model. You are provided with functions to execute. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools:
-    {{range .Functions}}
-    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
-    {{end}}
-    For each function call return a json object with function name and arguments
-    <|im_end|>
-    {{.Input -}}
-    <|im_start|>assistant
+    ASSISTANT:
 
 download_files:
-- filename: minicpm-v-2_6-Q4_K_M.gguf
-  sha256: 3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1
-  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/ggml-model-Q4_K_M.gguf
-- filename: minicpm-v-2_6-mmproj-f16.gguf
-  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/mmproj-model-f16.gguf
-  sha256: 4485f68a0f1aa404c391e788ea88ea653c100d8e98fe572698f701e5809711fd
+- filename: bakllava.gguf
+  uri: huggingface://mys/ggml_bakllava-1/ggml-model-q4_k.gguf
+- filename: bakllava-mmproj.gguf
+  uri: huggingface://mys/ggml_bakllava-1/mmproj-model-f16.gguf
+
+usage: |
+    curl http://localhost:8080/v1/chat/completions -H "Content-Type: application/json" -d '{
+        "model": "gpt-4-vision-preview",
+        "messages": [{"role": "user", "content": [{"type":"text", "text": "What is in the image?"}, {"type": "image_url", "image_url": {"url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg" }}], "temperature": 0.9}]}'

aio/gpu-8g/embeddings.yaml

@@ -1,7 +1,7 @@
-embeddings: true
 name: text-embedding-ada-002
+backend: sentencetransformers
 parameters:
-  model: huggingface://bartowski/granite-embedding-107m-multilingual-GGUF/granite-embedding-107m-multilingual-f16.gguf
+  model: all-MiniLM-L6-v2
 
 usage: |
     You can test this model with curl like this:

aio/gpu-8g/rerank.yaml

@@ -1,13 +1,7 @@
 name: jina-reranker-v1-base-en
-reranking: true
-f16: true
+backend: rerankers
 parameters:
-  model: jina-reranker-v1-tiny-en.f16.gguf
-
-download_files:
-  - filename: jina-reranker-v1-tiny-en.f16.gguf
-    sha256: 5f696cf0d0f3d347c4a279eee8270e5918554cdac0ed1f632f2619e4e8341407
-    uri: huggingface://mradermacher/jina-reranker-v1-tiny-en-GGUF/jina-reranker-v1-tiny-en.f16.gguf 
+  model: cross-encoder
 
 usage: |
     You can test this model with curl like this:

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

@@ -1,53 +1,101 @@
-context_size: 4096
-f16: true
-function:
-  capture_llm_results:
-  - (?s)<Thought>(.*?)</Thought>
-  grammar:
-    properties_order: name,arguments
-  json_regex_match:
-  - (?s)<Output>(.*?)</Output>
-  replace_llm_results:
-  - key: (?s)<Thought>(.*?)</Thought>
-    value: ""
-mmap: true
 name: gpt-4
+mmap: true
 parameters:
-  model: localai-functioncall-qwen2.5-7b-v0.5-q4_k_m.gguf
+  model: huggingface://NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+context_size: 8192
+
 stopwords:
-- <|im_end|>
-- <dummy32000>
-- </s>
+- "<|im_end|>"
+- "<dummy32000>"
+- "</tool_call>"
+- "<|eot_id|>"
+- "<|end_of_text|>"
+
+function:
+  # disable injecting the "answer" tool
+  disable_no_action: true
+
+  grammar:
+    # This allows the grammar to also return messages
+    mixed_mode: true
+    # Suffix to add to the grammar
+    #prefix: '<tool_call>\n'
+    # Force parallel calls in the grammar
+    # parallel_calls: true
+
+  return_name_in_function_response: true
+  # Without grammar uncomment the lines below
+  # Warning: this is relying only on the capability of the
+  # LLM model to generate the correct function call.
+  json_regex_match: 
+   - "(?s)<tool_call>(.*?)</tool_call>"
+   - "(?s)<tool_call>(.*?)"
+  replace_llm_results:
+  # Drop the scratchpad content from responses
+  - key: "(?s)<scratchpad>.*</scratchpad>"
+    value: ""
+  replace_function_results: 
+  # Replace everything that is not JSON array or object
+  # 
+  - key: '(?s)^[^{\[]*'
+    value: ""
+  - key: '(?s)[^}\]]*$'
+    value: ""
+  - key: "'([^']*?)'"
+    value: "_DQUOTE_${1}_DQUOTE_"
+  - key: '\\"'
+    value: "__TEMP_QUOTE__"
+  - key: "\'"
+    value: "'"
+  - key: "_DQUOTE_"
+    value: '"'
+  - key: "__TEMP_QUOTE__"
+    value: '"'
+  # Drop the scratchpad content from responses
+  - key: "(?s)<scratchpad>.*</scratchpad>"
+    value: ""
+
 template:
   chat: |
     {{.Input -}}
     <|im_start|>assistant
   chat_message: |
-    <|im_start|>{{ .RoleName }}
-    {{ if .FunctionCall -}}
-    Function call:
-    {{ else if eq .RoleName "tool" -}}
-    Function response:
-    {{ end -}}
-    {{ if .Content -}}
+    <|im_start|>{{if eq .RoleName "assistant"}}assistant{{else if eq .RoleName "system"}}system{{else if eq .RoleName "tool"}}tool{{else if eq .RoleName "user"}}user{{end}}
+    {{- if .FunctionCall }}
+    <tool_call>
+    {{- else if eq .RoleName "tool" }}
+    <tool_response>
+    {{- end }}
+    {{- if .Content}}
     {{.Content }}
-    {{ end -}}
-    {{ if .FunctionCall -}}
+    {{- end }}
+    {{- if .FunctionCall}}
     {{toJson .FunctionCall}}
-    {{ end -}}<|im_end|>
+    {{- end }}
+    {{- if .FunctionCall }}
+    </tool_call>
+    {{- else if eq .RoleName "tool" }}
+    </tool_response>
+    {{- end }}<|im_end|>
   completion: |
     {{.Input}}
-  function: |
+  function: |-
     <|im_start|>system
-    You are an AI assistant that executes function calls, and these are the tools at your disposal:
+    You are a function calling AI model.
+    Here are the available tools:
+    <tools>
     {{range .Functions}}
     {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
     {{end}}
-    <|im_end|>
+    </tools>
+    You should call the tools provided to you sequentially
+    Please use <scratchpad> XML tags to record your reasoning and planning before you call the functions as follows:
+    <scratchpad>
+    {step-by-step reasoning and plan in bullet points}
+    </scratchpad>
+    For each function call return a json object with function name and arguments within <tool_call> XML tags as follows:
+    <tool_call>
+    {"arguments": <args-dict>, "name": <function-name>}
+    </tool_call><|im_end|>
     {{.Input -}}
-    <|im_start|>assistant
-
-download_files:
-- filename: localai-functioncall-qwen2.5-7b-v0.5-q4_k_m.gguf
-  sha256: 4e7b7fe1d54b881f1ef90799219dc6cc285d29db24f559c8998d1addb35713d4
-  uri: huggingface://mudler/LocalAI-functioncall-qwen2.5-7b-v0.5-Q4_K_M-GGUF/localai-functioncall-qwen2.5-7b-v0.5-q4_k_m.gguf
+    <|im_start|>assistant

aio/gpu-8g/vision.yaml

@@ -1,49 +1,35 @@
+backend: llama-cpp
 context_size: 4096
 f16: true
 mmap: true
-mmproj: minicpm-v-2_6-mmproj-f16.gguf
 name: gpt-4o
+
+roles:
+  user: "USER:"
+  assistant: "ASSISTANT:"
+  system: "SYSTEM:"
+
+mmproj: llava-v1.6-7b-mmproj-f16.gguf
 parameters:
-  model: minicpm-v-2_6-Q4_K_M.gguf
-stopwords:
-- <|im_end|>
-- <dummy32000>
-- </s>
-- <|endoftext|>
+  model: llava-v1.6-mistral-7b.Q5_K_M.gguf
+  temperature: 0.2
+  top_k: 40
+  top_p: 0.95
+  seed: -1
+
 template:
   chat: |
-    {{.Input -}}
-    <|im_start|>assistant
-  chat_message: |
-    <|im_start|>{{ .RoleName }}
-    {{ if .FunctionCall -}}
-    Function call:
-    {{ else if eq .RoleName "tool" -}}
-    Function response:
-    {{ end -}}
-    {{ if .Content -}}
-    {{.Content }}
-    {{ end -}}
-    {{ if .FunctionCall -}}
-    {{toJson .FunctionCall}}
-    {{ end -}}<|im_end|>
-  completion: |
+    A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.
     {{.Input}}
-  function: |
-    <|im_start|>system
-    You are a function calling AI model. You are provided with functions to execute. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools:
-    {{range .Functions}}
-    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
-    {{end}}
-    For each function call return a json object with function name and arguments
-    <|im_end|>
-    {{.Input -}}
-    <|im_start|>assistant
+    ASSISTANT:
 
 download_files:
-- filename: minicpm-v-2_6-Q4_K_M.gguf
-  sha256: 3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1
-  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/ggml-model-Q4_K_M.gguf
-- filename: minicpm-v-2_6-mmproj-f16.gguf
-  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/mmproj-model-f16.gguf
-  sha256: 4485f68a0f1aa404c391e788ea88ea653c100d8e98fe572698f701e5809711fd
+- filename: llava-v1.6-mistral-7b.Q5_K_M.gguf
+  uri: huggingface://cjpais/llava-1.6-mistral-7b-gguf/llava-v1.6-mistral-7b.Q5_K_M.gguf
+- filename: llava-v1.6-7b-mmproj-f16.gguf
+  uri: huggingface://cjpais/llava-1.6-mistral-7b-gguf/mmproj-model-f16.gguf
+
+usage: |
+    curl http://localhost:8080/v1/chat/completions -H "Content-Type: application/json" -d '{
+        "model": "gpt-4-vision-preview",
+        "messages": [{"role": "user", "content": [{"type":"text", "text": "What is in the image?"}, {"type": "image_url", "image_url": {"url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg" }}], "temperature": 0.9}]}'

aio/intel/embeddings.yaml

@@ -1,7 +1,7 @@
-embeddings: true
 name: text-embedding-ada-002
+backend: sentencetransformers
 parameters:
-  model: huggingface://bartowski/granite-embedding-107m-multilingual-GGUF/granite-embedding-107m-multilingual-f16.gguf
+  model: all-MiniLM-L6-v2
 
 usage: |
     You can test this model with curl like this:

aio/intel/rerank.yaml

@@ -1,13 +1,7 @@
 name: jina-reranker-v1-base-en
-reranking: true
-f16: true
+backend: rerankers
 parameters:
-  model: jina-reranker-v1-tiny-en.f16.gguf
-
-download_files:
-  - filename: jina-reranker-v1-tiny-en.f16.gguf
-    sha256: 5f696cf0d0f3d347c4a279eee8270e5918554cdac0ed1f632f2619e4e8341407
-    uri: huggingface://mradermacher/jina-reranker-v1-tiny-en-GGUF/jina-reranker-v1-tiny-en.f16.gguf 
+  model: cross-encoder
 
 usage: |
     You can test this model with curl like this:

aio/intel/text-to-text.yaml

@@ -1,53 +1,103 @@
-context_size: 4096
-f16: true
-function:
-  capture_llm_results:
-  - (?s)<Thought>(.*?)</Thought>
-  grammar:
-    properties_order: name,arguments
-  json_regex_match:
-  - (?s)<Output>(.*?)</Output>
-  replace_llm_results:
-  - key: (?s)<Thought>(.*?)</Thought>
-    value: ""
-mmap: true
 name: gpt-4
+mmap: false
+context_size: 8192
+
+f16: false
 parameters:
-  model: localai-functioncall-qwen2.5-7b-v0.5-q4_k_m.gguf
+  model: huggingface://NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+
 stopwords:
-- <|im_end|>
-- <dummy32000>
-- </s>
+- "<|im_end|>"
+- "<dummy32000>"
+- "</tool_call>"
+- "<|eot_id|>"
+- "<|end_of_text|>"
+
+function:
+  # disable injecting the "answer" tool
+  disable_no_action: true
+
+  grammar:
+    # This allows the grammar to also return messages
+    mixed_mode: true
+    # Suffix to add to the grammar
+    #prefix: '<tool_call>\n'
+    # Force parallel calls in the grammar
+    # parallel_calls: true
+
+  return_name_in_function_response: true
+  # Without grammar uncomment the lines below
+  # Warning: this is relying only on the capability of the
+  # LLM model to generate the correct function call.
+  json_regex_match: 
+   - "(?s)<tool_call>(.*?)</tool_call>"
+   - "(?s)<tool_call>(.*?)"
+  replace_llm_results:
+  # Drop the scratchpad content from responses
+  - key: "(?s)<scratchpad>.*</scratchpad>"
+    value: ""
+  replace_function_results: 
+  # Replace everything that is not JSON array or object
+  # 
+  - key: '(?s)^[^{\[]*'
+    value: ""
+  - key: '(?s)[^}\]]*$'
+    value: ""
+  - key: "'([^']*?)'"
+    value: "_DQUOTE_${1}_DQUOTE_"
+  - key: '\\"'
+    value: "__TEMP_QUOTE__"
+  - key: "\'"
+    value: "'"
+  - key: "_DQUOTE_"
+    value: '"'
+  - key: "__TEMP_QUOTE__"
+    value: '"'
+  # Drop the scratchpad content from responses
+  - key: "(?s)<scratchpad>.*</scratchpad>"
+    value: ""
+
 template:
   chat: |
     {{.Input -}}
     <|im_start|>assistant
   chat_message: |
-    <|im_start|>{{ .RoleName }}
-    {{ if .FunctionCall -}}
-    Function call:
-    {{ else if eq .RoleName "tool" -}}
-    Function response:
-    {{ end -}}
-    {{ if .Content -}}
+    <|im_start|>{{if eq .RoleName "assistant"}}assistant{{else if eq .RoleName "system"}}system{{else if eq .RoleName "tool"}}tool{{else if eq .RoleName "user"}}user{{end}}
+    {{- if .FunctionCall }}
+    <tool_call>
+    {{- else if eq .RoleName "tool" }}
+    <tool_response>
+    {{- end }}
+    {{- if .Content}}
     {{.Content }}
-    {{ end -}}
-    {{ if .FunctionCall -}}
+    {{- end }}
+    {{- if .FunctionCall}}
     {{toJson .FunctionCall}}
-    {{ end -}}<|im_end|>
+    {{- end }}
+    {{- if .FunctionCall }}
+    </tool_call>
+    {{- else if eq .RoleName "tool" }}
+    </tool_response>
+    {{- end }}<|im_end|>
   completion: |
     {{.Input}}
-  function: |
+  function: |-
     <|im_start|>system
-    You are an AI assistant that executes function calls, and these are the tools at your disposal:
+    You are a function calling AI model.
+    Here are the available tools:
+    <tools>
     {{range .Functions}}
     {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
     {{end}}
-    <|im_end|>
+    </tools>
+    You should call the tools provided to you sequentially
+    Please use <scratchpad> XML tags to record your reasoning and planning before you call the functions as follows:
+    <scratchpad>
+    {step-by-step reasoning and plan in bullet points}
+    </scratchpad>
+    For each function call return a json object with function name and arguments within <tool_call> XML tags as follows:
+    <tool_call>
+    {"arguments": <args-dict>, "name": <function-name>}
+    </tool_call><|im_end|>
     {{.Input -}}
     <|im_start|>assistant
-
-download_files:
-- filename: localai-functioncall-phi-4-v0.3-q4_k_m.gguf
-  sha256: 23fee048ded2a6e2e1a7b6bbefa6cbf83068f194caa9552aecbaa00fec8a16d5
-  uri: huggingface://mudler/LocalAI-functioncall-phi-4-v0.3-Q4_K_M-GGUF/localai-functioncall-phi-4-v0.3-q4_k_m.gguf

aio/intel/vision.yaml

@@ -1,50 +1,35 @@
+backend: llama-cpp
 context_size: 4096
-f16: true
-mmap: true
-mmproj: minicpm-v-2_6-mmproj-f16.gguf
+mmap: false
+f16: false
 name: gpt-4o
+
+roles:
+  user: "USER:"
+  assistant: "ASSISTANT:"
+  system: "SYSTEM:"
+
+mmproj: llava-v1.6-7b-mmproj-f16.gguf
 parameters:
-  model: minicpm-v-2_6-Q4_K_M.gguf
-stopwords:
-- <|im_end|>
-- <dummy32000>
-- </s>
-- <|endoftext|>
+  model: llava-v1.6-mistral-7b.Q5_K_M.gguf
+  temperature: 0.2
+  top_k: 40
+  top_p: 0.95
+  seed: -1
+
 template:
   chat: |
-    {{.Input -}}
-    <|im_start|>assistant
-  chat_message: |
-    <|im_start|>{{ .RoleName }}
-    {{ if .FunctionCall -}}
-    Function call:
-    {{ else if eq .RoleName "tool" -}}
-    Function response:
-    {{ end -}}
-    {{ if .Content -}}
-    {{.Content }}
-    {{ end -}}
-    {{ if .FunctionCall -}}
-    {{toJson .FunctionCall}}
-    {{ end -}}<|im_end|>
-  completion: |
+    A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.
     {{.Input}}
-  function: |
-    <|im_start|>system
-    You are a function calling AI model. You are provided with functions to execute. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools:
-    {{range .Functions}}
-    {'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
-    {{end}}
-    For each function call return a json object with function name and arguments
-    <|im_end|>
-    {{.Input -}}
-    <|im_start|>assistant
-
+    ASSISTANT:
 
 download_files:
-- filename: minicpm-v-2_6-Q4_K_M.gguf
-  sha256: 3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1
-  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/ggml-model-Q4_K_M.gguf
-- filename: minicpm-v-2_6-mmproj-f16.gguf
-  uri: huggingface://openbmb/MiniCPM-V-2_6-gguf/mmproj-model-f16.gguf
-  sha256: 4485f68a0f1aa404c391e788ea88ea653c100d8e98fe572698f701e5809711fd
+- filename: llava-v1.6-mistral-7b.Q5_K_M.gguf
+  uri: huggingface://cjpais/llava-1.6-mistral-7b-gguf/llava-v1.6-mistral-7b.Q5_K_M.gguf
+- filename: llava-v1.6-7b-mmproj-f16.gguf
+  uri: huggingface://cjpais/llava-1.6-mistral-7b-gguf/mmproj-model-f16.gguf
+
+usage: |
+    curl http://localhost:8080/v1/chat/completions -H "Content-Type: application/json" -d '{
+        "model": "gpt-4-vision-preview",
+        "messages": [{"role": "user", "content": [{"type":"text", "text": "What is in the image?"}, {"type": "image_url", "image_url": {"url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg" }}], "temperature": 0.9}]}'

assets.go

@@ -1,15 +1,6 @@
 package main
 
-import (
-	rice "github.com/GeertJohan/go.rice"
-)
+import "embed"
 
-var backendAssets *rice.Box
-
-func init() {
-	var err error
-	backendAssets, err = rice.FindBox("backend-assets")
-	if err != nil {
-		panic(err)
-	}
-}
+//go:embed backend-assets/*
+var backendAssets embed.FS

backend/Dockerfile.go

@@ -1,131 +0,0 @@
-ARG BASE_IMAGE=ubuntu:22.04
-
-FROM ${BASE_IMAGE} AS builder
-ARG BACKEND=rerankers
-ARG BUILD_TYPE
-ENV BUILD_TYPE=${BUILD_TYPE}
-ARG CUDA_MAJOR_VERSION
-ARG CUDA_MINOR_VERSION
-ARG SKIP_DRIVERS=false
-ENV CUDA_MAJOR_VERSION=${CUDA_MAJOR_VERSION}
-ENV CUDA_MINOR_VERSION=${CUDA_MINOR_VERSION}
-ENV DEBIAN_FRONTEND=noninteractive
-ARG TARGETARCH
-ARG TARGETVARIANT
-ARG GO_VERSION=1.22.6
-
-RUN apt-get update && \
-    apt-get install -y --no-install-recommends \
-        build-essential \
-        ccache \
-        ca-certificates \
-        make \
-        curl unzip \
-        libssl-dev && \
-    apt-get clean && \
-    rm -rf /var/lib/apt/lists/*
-
-
-# Cuda
-ENV PATH=/usr/local/cuda/bin:${PATH}
-
-# HipBLAS requirements
-ENV PATH=/opt/rocm/bin:${PATH}
-
-# Vulkan requirements
-RUN <<EOT bash
-    if [ "${BUILD_TYPE}" = "vulkan" ] && [ "${SKIP_DRIVERS}" = "false" ]; then
-        apt-get update && \
-        apt-get install -y  --no-install-recommends \
-            software-properties-common pciutils wget gpg-agent && \
-        wget -qO - https://packages.lunarg.com/lunarg-signing-key-pub.asc | apt-key add - && \
-        wget -qO /etc/apt/sources.list.d/lunarg-vulkan-jammy.list https://packages.lunarg.com/vulkan/lunarg-vulkan-jammy.list && \
-        apt-get update && \
-        apt-get install -y \
-            vulkan-sdk && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/*
-    fi
-EOT
-
-# CuBLAS requirements
-RUN <<EOT bash
-    if [ "${BUILD_TYPE}" = "cublas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then
-        apt-get update && \
-        apt-get install -y  --no-install-recommends \
-            software-properties-common pciutils
-        if [ "amd64" = "$TARGETARCH" ]; then
-            curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
-        fi
-        if [ "arm64" = "$TARGETARCH" ]; then
-            curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/arm64/cuda-keyring_1.1-1_all.deb
-        fi
-        dpkg -i cuda-keyring_1.1-1_all.deb && \
-        rm -f cuda-keyring_1.1-1_all.deb && \
-        apt-get update && \
-        apt-get install -y --no-install-recommends \
-            cuda-nvcc-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcufft-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcurand-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcublas-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcusparse-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcusolver-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/*
-    fi
-EOT
-
-# If we are building with clblas support, we need the libraries for the builds
-RUN if [ "${BUILD_TYPE}" = "clblas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then \
-        apt-get update && \
-        apt-get install -y --no-install-recommends \
-            libclblast-dev && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/* \
-    ; fi
-
-RUN if [ "${BUILD_TYPE}" = "hipblas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then \
-        apt-get update && \
-        apt-get install -y --no-install-recommends \
-            hipblas-dev \
-            rocblas-dev && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/* && \
-        # I have no idea why, but the ROCM lib packages don't trigger ldconfig after they install, which results in local-ai and others not being able
-        # to locate the libraries. We run ldconfig ourselves to work around this packaging deficiency
-        ldconfig \
-    ; fi
-
-# Install Go
-RUN curl -L -s https://go.dev/dl/go${GO_VERSION}.linux-${TARGETARCH}.tar.gz | tar -C /usr/local -xz
-ENV PATH=$PATH:/root/go/bin:/usr/local/go/bin:/usr/local/bin
-
-# Install grpc compilers
-RUN go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2 && \
-    go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
-RUN echo "TARGETARCH: $TARGETARCH"
-
-# We need protoc installed, and the version in 22.04 is too old.  We will create one as part installing the GRPC build below
-# but that will also being in a newer version of absl which stablediffusion cannot compile with.  This version of protoc is only
-# here so that we can generate the grpc code for the stablediffusion build
-RUN <<EOT bash
-    if [ "amd64" = "$TARGETARCH" ]; then
-        curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v27.1/protoc-27.1-linux-x86_64.zip -o protoc.zip && \
-        unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
-        rm protoc.zip
-    fi
-    if [ "arm64" = "$TARGETARCH" ]; then
-        curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v27.1/protoc-27.1-linux-aarch_64.zip -o protoc.zip && \
-        unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
-        rm protoc.zip
-    fi
-EOT
-
-COPY . /LocalAI
-
-RUN cd /LocalAI && make backend-assets/grpc/bark-cpp
-
-FROM scratch
-
-COPY --from=builder /LocalAI/backend-assets/grpc/bark-cpp ./
-COPY --from=builder /LocalAI/backend/go/bark/run.sh ./

backend/Dockerfile.python

@@ -1,123 +0,0 @@
-ARG BASE_IMAGE=ubuntu:22.04
-
-FROM ${BASE_IMAGE} AS builder
-ARG BACKEND=rerankers
-ARG BUILD_TYPE
-ENV BUILD_TYPE=${BUILD_TYPE}
-ARG CUDA_MAJOR_VERSION
-ARG CUDA_MINOR_VERSION
-ARG SKIP_DRIVERS=false
-ENV CUDA_MAJOR_VERSION=${CUDA_MAJOR_VERSION}
-ENV CUDA_MINOR_VERSION=${CUDA_MINOR_VERSION}
-ENV DEBIAN_FRONTEND=noninteractive
-ARG TARGETARCH
-ARG TARGETVARIANT
-
-RUN apt-get update && \
-    apt-get install -y --no-install-recommends \
-        build-essential \
-        ccache \
-        ca-certificates \
-        espeak-ng \
-        curl \
-        libssl-dev \
-        git \
-        git-lfs \
-        unzip \
-        upx-ucl \
-        curl python3-pip \
-        python-is-python3 \
-        python3-dev llvm \
-        python3-venv make && \
-    apt-get clean && \
-    rm -rf /var/lib/apt/lists/* && \
-    pip install --upgrade pip
-
-
-# Cuda
-ENV PATH=/usr/local/cuda/bin:${PATH}
-
-# HipBLAS requirements
-ENV PATH=/opt/rocm/bin:${PATH}
-
-# Vulkan requirements
-RUN <<EOT bash
-    if [ "${BUILD_TYPE}" = "vulkan" ] && [ "${SKIP_DRIVERS}" = "false" ]; then
-        apt-get update && \
-        apt-get install -y  --no-install-recommends \
-            software-properties-common pciutils wget gpg-agent && \
-        wget -qO - https://packages.lunarg.com/lunarg-signing-key-pub.asc | apt-key add - && \
-        wget -qO /etc/apt/sources.list.d/lunarg-vulkan-jammy.list https://packages.lunarg.com/vulkan/lunarg-vulkan-jammy.list && \
-        apt-get update && \
-        apt-get install -y \
-            vulkan-sdk && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/*
-    fi
-EOT
-
-# CuBLAS requirements
-RUN <<EOT bash
-    if [ "${BUILD_TYPE}" = "cublas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then
-        apt-get update && \
-        apt-get install -y  --no-install-recommends \
-            software-properties-common pciutils
-        if [ "amd64" = "$TARGETARCH" ]; then
-            curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
-        fi
-        if [ "arm64" = "$TARGETARCH" ]; then
-            curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/arm64/cuda-keyring_1.1-1_all.deb
-        fi
-        dpkg -i cuda-keyring_1.1-1_all.deb && \
-        rm -f cuda-keyring_1.1-1_all.deb && \
-        apt-get update && \
-        apt-get install -y --no-install-recommends \
-            cuda-nvcc-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcufft-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcurand-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcublas-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcusparse-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcusolver-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/*
-    fi
-EOT
-
-# If we are building with clblas support, we need the libraries for the builds
-RUN if [ "${BUILD_TYPE}" = "clblas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then \
-        apt-get update && \
-        apt-get install -y --no-install-recommends \
-            libclblast-dev && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/* \
-    ; fi
-
-RUN if [ "${BUILD_TYPE}" = "hipblas" ] && [ "${SKIP_DRIVERS}" = "false" ]; then \
-        apt-get update && \
-        apt-get install -y --no-install-recommends \
-            hipblas-dev \
-            rocblas-dev && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/* && \
-        # I have no idea why, but the ROCM lib packages don't trigger ldconfig after they install, which results in local-ai and others not being able
-        # to locate the libraries. We run ldconfig ourselves to work around this packaging deficiency
-        ldconfig \
-    ; fi
-# Install uv as a system package
-RUN curl -LsSf https://astral.sh/uv/install.sh | UV_INSTALL_DIR=/usr/bin sh
-ENV PATH="/root/.cargo/bin:${PATH}"
-
-RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
-
-# Install grpcio-tools (the version in 22.04 is too old)
-RUN pip install --user grpcio-tools==1.71.0 grpcio==1.71.0
-
-COPY python/${BACKEND} /${BACKEND}
-COPY backend.proto /${BACKEND}/backend.proto
-COPY python/common/ /${BACKEND}/common
-
-RUN cd /${BACKEND} && make
-
-FROM scratch
-ARG BACKEND=rerankers
-COPY --from=builder /${BACKEND}/ /

backend/backend.proto

@@ -14,7 +14,6 @@ service Backend {
   rpc PredictStream(PredictOptions) returns (stream Reply) {}
   rpc Embedding(PredictOptions) returns (EmbeddingResult) {}
   rpc GenerateImage(GenerateImageRequest) returns (Result) {}
-  rpc GenerateVideo(GenerateVideoRequest) returns (Result) {}
   rpc AudioTranscription(TranscriptRequest) returns (TranscriptResult) {}
   rpc TTS(TTSRequest) returns (Result) {}
   rpc SoundGeneration(SoundGenerationRequest) returns (Result) {}
@@ -162,7 +161,6 @@ message Reply {
   int32 prompt_tokens = 3;
   double timing_prompt_processing = 4;
   double timing_token_generation = 5;
-  bytes audio = 6;
 }
 
 message GrammarTrigger {
@@ -192,7 +190,11 @@ message ModelOptions {
   int32 NGQA = 20;
   string ModelFile = 21;
 
-
+  // AutoGPTQ
+  string Device = 22;
+  bool UseTriton = 23;
+  string ModelBaseName = 24;
+  bool UseFastTokenizer = 25;
 
   // Diffusers
   string PipelineType = 26;
@@ -256,8 +258,6 @@ message ModelOptions {
   string CacheTypeValue = 64;
 
   repeated GrammarTrigger GrammarTriggers = 65;
-
-  bool Reranking = 71;
 }
 
 message Result {
@@ -305,19 +305,6 @@ message GenerateImageRequest {
   int32 CLIPSkip = 11;
 }
 
-message GenerateVideoRequest {
-  string prompt = 1;
-  string start_image = 2;  // Path or base64 encoded image for the start frame
-  string end_image = 3;    // Path or base64 encoded image for the end frame
-  int32 width = 4;
-  int32 height = 5;
-  int32 num_frames = 6;    // Number of frames to generate
-  int32 fps = 7;          // Frames per second
-  int32 seed = 8;
-  float cfg_scale = 9;    // Classifier-free guidance scale
-  string dst = 10;        // Output path for the generated video
-}
-
 message TTSRequest {
   string text = 1;
   string model = 2;

backend/cpp/llama/CMakeLists.txt

@@ -1,3 +1,20 @@
+
+## XXX: In some versions of CMake clip wasn't being built before llama.
+## This is an hack for now, but it should be fixed in the future.
+set(TARGET myclip)
+add_library(${TARGET} clip.cpp clip.h llava.cpp llava.h)
+install(TARGETS ${TARGET} LIBRARY)
+target_include_directories(myclip PUBLIC .)
+target_include_directories(myclip PUBLIC ../..)
+target_include_directories(myclip PUBLIC ../../common)
+target_link_libraries(${TARGET} PRIVATE common ggml llama ${CMAKE_THREAD_LIBS_INIT})
+target_compile_features(${TARGET} PRIVATE cxx_std_11)
+if (NOT MSVC)
+    target_compile_options(${TARGET} PRIVATE -Wno-cast-qual) # stb_image.h
+endif()
+# END CLIP hack
+
+
 set(TARGET grpc-server)
 set(CMAKE_CXX_STANDARD 17)
 cmake_minimum_required(VERSION 3.15)
@@ -57,12 +74,8 @@ add_library(hw_grpc_proto
   ${hw_proto_srcs}
   ${hw_proto_hdrs} )
 
-add_executable(${TARGET} grpc-server.cpp utils.hpp json.hpp httplib.h)
-
-target_include_directories(${TARGET} PRIVATE ../llava)
-target_include_directories(${TARGET} PRIVATE ${CMAKE_SOURCE_DIR})
-
-target_link_libraries(${TARGET} PRIVATE common llama mtmd ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
+add_executable(${TARGET} grpc-server.cpp utils.hpp json.hpp)
+target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
   absl::flags_parse
   gRPC::${_REFLECTION}
   gRPC::${_GRPC_GRPCPP}

backend/cpp/llama/Makefile

@@ -8,7 +8,7 @@ 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 -DLLAMA_CURL=OFF
+CMAKE_ARGS+=-DBUILD_SHARED_LIBS=OFF
 
 # If build type is cublas, then we set -DGGML_CUDA=ON to CMAKE_ARGS automatically
 ifeq ($(BUILD_TYPE),cublas)
@@ -36,18 +36,11 @@ else ifeq ($(OS),Darwin)
 endif
 
 ifeq ($(BUILD_TYPE),sycl_f16)
-	CMAKE_ARGS+=-DGGML_SYCL=ON \
-		-DCMAKE_C_COMPILER=icx \
-		-DCMAKE_CXX_COMPILER=icpx \
-		-DCMAKE_CXX_FLAGS="-fsycl" \
-		-DGGML_SYCL_F16=ON
+	CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DGGML_SYCL_F16=ON
 endif
 
 ifeq ($(BUILD_TYPE),sycl_f32)
-	CMAKE_ARGS+=-DGGML_SYCL=ON \
-		-DCMAKE_C_COMPILER=icx \
-		-DCMAKE_CXX_COMPILER=icpx \
-		-DCMAKE_CXX_FLAGS="-fsycl"
+	CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
 endif
 
 llama.cpp:
@@ -59,8 +52,8 @@ llama.cpp:
 	git checkout -b build $(LLAMA_VERSION) && \
 	git submodule update --init --recursive --depth 1 --single-branch
 
-llama.cpp/tools/grpc-server: llama.cpp
-	mkdir -p llama.cpp/tools/grpc-server
+llama.cpp/examples/grpc-server: llama.cpp
+	mkdir -p llama.cpp/examples/grpc-server
 	bash prepare.sh
 
 rebuild:
@@ -70,13 +63,13 @@ rebuild:
 
 purge:
 	rm -rf llama.cpp/build
-	rm -rf llama.cpp/tools/grpc-server
+	rm -rf llama.cpp/examples/grpc-server
 	rm -rf grpc-server
 
 clean: purge
 	rm -rf llama.cpp
 
-grpc-server: llama.cpp llama.cpp/tools/grpc-server
+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 $(ONEAPI_VARS); \
@@ -84,4 +77,4 @@ ifneq (,$(findstring sycl,$(BUILD_TYPE)))
 else
 	+cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && cmake --build . --config Release $(TARGET)
 endif
-	cp llama.cpp/build/bin/grpc-server .
+	cp llama.cpp/build/bin/grpc-server .

backend/cpp/llama/patches/01-llava.patch

@@ -1,13 +1,13 @@
-diff --git a/tools/mtmd/clip.cpp b/tools/mtmd/clip.cpp
-index 3cd0d2fa..6c5e811a 100644
---- a/tools/mtmd/clip.cpp
-+++ b/tools/mtmd/clip.cpp
-@@ -2608,7 +2608,7 @@ bool clip_image_batch_encode(clip_ctx * ctx, const int n_threads, const clip_ima
-                 struct ggml_tensor * patches = ggml_graph_get_tensor(gf, "patches");
+diff --git a/examples/llava/clip.cpp b/examples/llava/clip.cpp
+index 7f892beb..0517e529 100644
+--- a/examples/llava/clip.cpp
++++ b/examples/llava/clip.cpp
+@@ -2766,7 +2766,7 @@ bool clip_image_batch_encode(clip_ctx * ctx, const int n_threads, const clip_ima
+                 int patch_offset = ctx->has_class_embedding ? 1 : 0;
                  int* patches_data = (int*)malloc(ggml_nbytes(patches));
                  for (int i = 0; i < num_patches; i++) {
--                    patches_data[i] = i + 1;
-+                    patches_data[i] = i;
+-                    patches_data[i] = i + patch_offset;
++                    patches_data[i] = i + 1;
                  }
                  ggml_backend_tensor_set(patches, patches_data, 0, ggml_nbytes(patches));
                  free(patches_data);

backend/cpp/llama/prepare.sh

@@ -1,5 +1,7 @@
 #!/bin/bash
 
+set -e
+
 ## Patches
 ## Apply patches from the `patches` directory
 for patch in $(ls patches); do
@@ -7,46 +9,21 @@ for patch in $(ls patches); do
     patch -d llama.cpp/ -p1 < patches/$patch
 done 
 
-set -e
-
-cp -r CMakeLists.txt llama.cpp/tools/grpc-server/
-cp -r grpc-server.cpp llama.cpp/tools/grpc-server/
-cp -rfv llama.cpp/vendor/nlohmann/json.hpp llama.cpp/tools/grpc-server/
-cp -rfv llama.cpp/tools/server/utils.hpp llama.cpp/tools/grpc-server/
-cp -rfv llama.cpp/vendor/cpp-httplib/httplib.h llama.cpp/tools/grpc-server/
-
-set +e
-if grep -q "grpc-server" llama.cpp/tools/CMakeLists.txt; then
+cp -r CMakeLists.txt llama.cpp/examples/grpc-server/
+cp -r grpc-server.cpp llama.cpp/examples/grpc-server/
+cp -rfv json.hpp llama.cpp/examples/grpc-server/
+cp -rfv utils.hpp llama.cpp/examples/grpc-server/
+    
+if grep -q "grpc-server" llama.cpp/examples/CMakeLists.txt; then
     echo "grpc-server already added"
 else
-    echo "add_subdirectory(grpc-server)" >> llama.cpp/tools/CMakeLists.txt
+    echo "add_subdirectory(grpc-server)" >> llama.cpp/examples/CMakeLists.txt
 fi
-set -e
 
-# Now to keep maximum compatibility with the original server.cpp, we need to remove the index.html.gz.hpp and loading.html.hpp includes
-# and remove the main function
-# TODO: upstream this to the original server.cpp by extracting the upstream main function to a separate file
-awk '
-/int[ \t]+main[ \t]*\(/ {          # If the line starts the main function
-    in_main=1;                     # Set a flag
-    open_braces=0;                 # Track number of open braces
-}
-in_main {
-    open_braces += gsub(/\{/, "{"); # Count opening braces
-    open_braces -= gsub(/\}/, "}"); # Count closing braces
-    if (open_braces == 0) {         # If all braces are closed
-        in_main=0;                  # End skipping
-    }
-    next;                           # Skip lines inside main
-}
-!in_main                           # Print lines not inside main
-' "llama.cpp/tools/server/server.cpp" > llama.cpp/tools/grpc-server/server.cpp
-
-# remove index.html.gz.hpp and loading.html.hpp includes
-if [[ "$OSTYPE" == "darwin"* ]]; then
-    # macOS
-    sed -i '' '/#include "index\.html\.gz\.hpp"/d; /#include "loading\.html\.hpp"/d' llama.cpp/tools/grpc-server/server.cpp
-else
-    # Linux and others
-    sed -i '/#include "index\.html\.gz\.hpp"/d; /#include "loading\.html\.hpp"/d' llama.cpp/tools/grpc-server/server.cpp
-fi
+## XXX: In some versions of CMake clip wasn't being built before llama.
+## This is an hack for now, but it should be fixed in the future.
+cp -rfv llama.cpp/examples/llava/clip.h llama.cpp/examples/grpc-server/clip.h
+cp -rfv llama.cpp/examples/llava/llava.cpp llama.cpp/examples/grpc-server/llava.cpp
+echo '#include "llama.h"' > llama.cpp/examples/grpc-server/llava.h
+cat llama.cpp/examples/llava/llava.h >> llama.cpp/examples/grpc-server/llava.h
+cp -rfv llama.cpp/examples/llava/clip.cpp llama.cpp/examples/grpc-server/clip.cpp

backend/cpp/llama/utils.hpp

@@ -0,0 +1,483 @@
+// https://github.com/ggerganov/llama.cpp/blob/master/examples/server/utils.hpp
+
+#pragma once
+
+#include <string>
+#include <vector>
+#include <set>
+#include <mutex>
+#include <condition_variable>
+#include <unordered_map>
+
+#include "json.hpp"
+
+#include "../llava/clip.h"
+
+using json = nlohmann::json;
+
+extern bool server_verbose;
+
+#ifndef SERVER_VERBOSE
+#define SERVER_VERBOSE 1
+#endif
+
+#if SERVER_VERBOSE != 1
+#define LOG_VERBOSE(MSG, ...)
+#else
+#define LOG_VERBOSE(MSG, ...)                                            \
+    do                                                                   \
+    {                                                                    \
+        if (server_verbose)                                              \
+        {                                                                \
+            server_log("VERBOSE", __func__, __LINE__, MSG, __VA_ARGS__); \
+        }                                                                \
+    } while (0)
+#endif
+
+#define LOG_ERROR(  MSG, ...) server_log("ERROR",   __func__, __LINE__, MSG, __VA_ARGS__)
+#define LOG_WARNING(MSG, ...) server_log("WARNING", __func__, __LINE__, MSG, __VA_ARGS__)
+#define LOG_INFO(   MSG, ...) server_log("INFO",    __func__, __LINE__, MSG, __VA_ARGS__)
+
+//
+// parallel
+//
+
+enum server_state {
+    SERVER_STATE_LOADING_MODEL,  // Server is starting up, model not fully loaded yet
+    SERVER_STATE_READY,          // Server is ready and model is loaded
+    SERVER_STATE_ERROR           // An error occurred, load_model failed
+};
+
+enum task_type {
+    TASK_TYPE_COMPLETION,
+    TASK_TYPE_CANCEL,
+    TASK_TYPE_NEXT_RESPONSE
+};
+
+struct task_server {
+    int id = -1; // to be filled by llama_server_queue
+    int target_id;
+    task_type type;
+    json data;
+    bool infill_mode = false;
+    bool embedding_mode = false;
+    int multitask_id = -1;
+};
+
+struct task_result {
+    int id;
+    int multitask_id = -1;
+    bool stop;
+    bool error;
+    json result_json;
+};
+
+struct task_multi {
+    int id;
+    std::set<int> subtasks_remaining{};
+    std::vector<task_result> results{};
+};
+
+// TODO: can become bool if we can't find use of more states
+enum slot_state
+{
+    IDLE,
+    PROCESSING,
+};
+
+enum slot_command
+{
+    NONE,
+    LOAD_PROMPT,
+    RELEASE,
+};
+
+struct slot_params
+{
+    bool stream       = true;
+    bool cache_prompt = false; // remember the prompt to avoid reprocessing all prompt
+
+    uint32_t seed      = -1; // RNG seed
+    int32_t  n_keep    =  0; // number of tokens to keep from initial prompt
+    int32_t  n_predict = -1; // new tokens to predict
+
+    std::vector<std::string> antiprompt;
+
+    json input_prefix;
+    json input_suffix;
+};
+
+struct slot_image
+{
+    int32_t id;
+
+    bool request_encode_image = false;
+    float * image_embedding = nullptr;
+    int32_t image_tokens = 0;
+
+    clip_image_u8 * img_data;
+
+    std::string prefix_prompt; // before of this image
+};
+
+// completion token output with probabilities
+struct completion_token_output
+{
+    struct token_prob
+    {
+        llama_token tok;
+        float prob;
+    };
+
+    std::vector<token_prob> probs;
+    llama_token tok;
+    std::string text_to_send;
+};
+
+static inline void server_log(const char *level, const char *function, int line,
+                       const char *message, const nlohmann::ordered_json &extra)
+{
+    nlohmann::ordered_json log
+    {
+        {"timestamp", time(nullptr)},
+        {"level",     level},
+        {"function",  function},
+        {"line",      line},
+        {"message",   message},
+    };
+
+    if (!extra.empty())
+    {
+        log.merge_patch(extra);
+    }
+
+    const std::string str = log.dump(-1, ' ', false, json::error_handler_t::replace);
+    printf("%.*s\n", (int)str.size(), str.data());
+    fflush(stdout);
+}
+
+//
+// server utils
+//
+
+template <typename T>
+static T json_value(const json &body, const std::string &key, const T &default_value)
+{
+    // Fallback null to default value
+    return body.contains(key) && !body.at(key).is_null()
+        ? body.value(key, default_value)
+        : default_value;
+}
+
+inline std::string format_chatml(std::vector<json> messages)
+{
+    std::ostringstream chatml_msgs;
+
+    for (auto it = messages.begin(); it != messages.end(); ++it) {
+        chatml_msgs << "<|im_start|>"
+                    << json_value(*it, "role",    std::string("user")) << '\n';
+        chatml_msgs << json_value(*it, "content", std::string(""))
+                    << "<|im_end|>\n";
+    }
+
+    chatml_msgs << "<|im_start|>assistant" << '\n';
+
+    return chatml_msgs.str();
+}
+
+//
+// work queue utils
+//
+
+struct llama_server_queue {
+    int id = 0;
+    std::mutex mutex_tasks;
+    // queues
+    std::vector<task_server> queue_tasks;
+    std::vector<task_server> queue_tasks_deferred;
+    std::vector<task_multi> queue_multitasks;
+    std::condition_variable condition_tasks;
+    // callback functions
+    std::function<void(task_server&)> callback_new_task;
+    std::function<void(task_multi&)> callback_finish_multitask;
+    std::function<void(void)> callback_all_task_finished;
+
+    // Add a new task to the end of the queue
+    int post(task_server task) {
+        std::unique_lock<std::mutex> lock(mutex_tasks);
+        if (task.id == -1) {
+            task.id = id++;
+        }
+        queue_tasks.push_back(std::move(task));
+        condition_tasks.notify_one();
+        return task.id;
+    }
+
+    // Add a new task, but defer until one slot is available
+    void defer(task_server task) {
+        std::unique_lock<std::mutex> lock(mutex_tasks);
+        queue_tasks_deferred.push_back(std::move(task));
+    }
+
+    // Get the next id for creating anew task
+    int get_new_id() {
+        std::unique_lock<std::mutex> lock(mutex_tasks);
+        return id++;
+    }
+
+    // Register function to process a new task
+    void on_new_task(std::function<void(task_server&)> callback) {
+        callback_new_task = callback;
+    }
+
+    // Register function to process a multitask
+    void on_finish_multitask(std::function<void(task_multi&)> callback) {
+        callback_finish_multitask = callback;
+    }
+
+    // Register the function to be called when the batch of tasks is finished
+    void on_all_tasks_finished(std::function<void(void)> callback) {
+        callback_all_task_finished = callback;
+    }
+
+    // Call when the state of one slot is changed
+    void notify_slot_changed() {
+        // move deferred tasks back to main loop
+        std::unique_lock<std::mutex> lock(mutex_tasks);
+        for (auto & task : queue_tasks_deferred) {
+            queue_tasks.push_back(std::move(task));
+        }
+        queue_tasks_deferred.clear();
+    }
+
+    // Start the main loop. This call is blocking
+    [[noreturn]]
+    void start_loop() {
+        while (true) {
+            // new task arrived
+            LOG_VERBOSE("have new task", {});
+            {
+                while (true)
+                {
+                    std::unique_lock<std::mutex> lock(mutex_tasks);
+                    if (queue_tasks.empty()) {
+                        lock.unlock();
+                        break;
+                    }
+                    task_server task = queue_tasks.front();
+                    queue_tasks.erase(queue_tasks.begin());
+                    lock.unlock();
+                    LOG_VERBOSE("callback_new_task", {});
+                    callback_new_task(task);
+                }
+                LOG_VERBOSE("callback_all_task_finished", {});
+                // process and update all the multitasks
+                auto queue_iterator = queue_multitasks.begin();
+                while (queue_iterator != queue_multitasks.end())
+                {
+                    if (queue_iterator->subtasks_remaining.empty())
+                    {
+                        // all subtasks done == multitask is done
+                        task_multi current_multitask = *queue_iterator;
+                        callback_finish_multitask(current_multitask);
+                        // remove this multitask
+                        queue_iterator = queue_multitasks.erase(queue_iterator);
+                    }
+                    else
+                    {
+                        ++queue_iterator;
+                    }
+                }
+                // all tasks in the current loop is finished
+                callback_all_task_finished();
+            }
+            LOG_VERBOSE("wait for new task", {});
+            // wait for new task
+            {
+                std::unique_lock<std::mutex> lock(mutex_tasks);
+                if (queue_tasks.empty()) {
+                    condition_tasks.wait(lock, [&]{
+                        return !queue_tasks.empty();
+                    });
+                }
+            }
+        }
+    }
+
+    //
+    // functions to manage multitasks
+    //
+
+    // add a multitask by specifying the id of all subtask (subtask is a task_server)
+    void add_multitask(int multitask_id, std::vector<int>& sub_ids)
+    {
+        std::lock_guard<std::mutex> lock(mutex_tasks);
+        task_multi multi;
+        multi.id = multitask_id;
+        std::copy(sub_ids.begin(), sub_ids.end(), std::inserter(multi.subtasks_remaining, multi.subtasks_remaining.end()));
+        queue_multitasks.push_back(multi);
+    }
+
+    // updatethe remaining subtasks, while appending results to multitask
+    void update_multitask(int multitask_id, int subtask_id, task_result& result)
+    {
+        std::lock_guard<std::mutex> lock(mutex_tasks);
+        for (auto& multitask : queue_multitasks)
+        {
+            if (multitask.id == multitask_id)
+            {
+                multitask.subtasks_remaining.erase(subtask_id);
+                multitask.results.push_back(result);
+            }
+        }
+    }
+};
+
+struct llama_server_response {
+    typedef std::function<void(int, int, task_result&)> callback_multitask_t;
+    callback_multitask_t callback_update_multitask;
+    // for keeping track of all tasks waiting for the result
+    std::set<int> waiting_task_ids;
+    // the main result queue
+    std::vector<task_result> queue_results;
+    std::mutex mutex_results;
+    std::condition_variable condition_results;
+
+    void add_waiting_task_id(int task_id) {
+        std::unique_lock<std::mutex> lock(mutex_results);
+        waiting_task_ids.insert(task_id);
+    }
+
+    void remove_waiting_task_id(int task_id) {
+        std::unique_lock<std::mutex> lock(mutex_results);
+        waiting_task_ids.erase(task_id);
+    }
+
+    // This function blocks the thread until there is a response for this task_id
+    task_result recv(int task_id) {
+        while (true)
+        {
+            std::unique_lock<std::mutex> lock(mutex_results);
+            condition_results.wait(lock, [&]{
+                return !queue_results.empty();
+            });
+            LOG_VERBOSE("condition_results unblock", {});
+
+            for (int i = 0; i < (int) queue_results.size(); i++)
+            {
+                if (queue_results[i].id == task_id)
+                {
+                    assert(queue_results[i].multitask_id == -1);
+                    task_result res = queue_results[i];
+                    queue_results.erase(queue_results.begin() + i);
+                    return res;
+                }
+            }
+        }
+
+        // should never reach here
+    }
+
+    // Register the function to update multitask
+    void on_multitask_update(callback_multitask_t callback) {
+        callback_update_multitask = callback;
+    }
+
+    // Send a new result to a waiting task_id
+    void send(task_result result) {
+        std::unique_lock<std::mutex> lock(mutex_results);
+        LOG_VERBOSE("send new result", {});
+        for (auto& task_id : waiting_task_ids) {
+            // LOG_TEE("waiting task id %i \n", task_id);
+            // for now, tasks that have associated parent multitasks just get erased once multitask picks up the result
+            if (result.multitask_id == task_id)
+            {
+                LOG_VERBOSE("callback_update_multitask", {});
+                callback_update_multitask(task_id, result.id, result);
+                continue;
+            }
+
+            if (result.id == task_id)
+            {
+                LOG_VERBOSE("queue_results.push_back", {});
+                queue_results.push_back(result);
+                condition_results.notify_one();
+                return;
+            }
+        }
+    }
+};
+
+//
+// base64 utils (TODO: move to common in the future)
+//
+
+static const std::string base64_chars =
+             "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
+             "abcdefghijklmnopqrstuvwxyz"
+             "0123456789+/";
+
+static inline bool is_base64(uint8_t c)
+{
+    return (isalnum(c) || (c == '+') || (c == '/'));
+}
+
+static inline std::vector<uint8_t> base64_decode(const std::string & encoded_string)
+{
+    int i = 0;
+    int j = 0;
+    int in_ = 0;
+
+    int in_len = encoded_string.size();
+
+    uint8_t char_array_4[4];
+    uint8_t char_array_3[3];
+
+    std::vector<uint8_t> ret;
+
+    while (in_len-- && (encoded_string[in_] != '=') && is_base64(encoded_string[in_]))
+    {
+        char_array_4[i++] = encoded_string[in_]; in_++;
+        if (i == 4)
+        {
+            for (i = 0; i <4; i++)
+            {
+                char_array_4[i] = base64_chars.find(char_array_4[i]);
+            }
+
+            char_array_3[0] = ((char_array_4[0]      ) << 2) + ((char_array_4[1] & 0x30) >> 4);
+            char_array_3[1] = ((char_array_4[1] & 0xf) << 4) + ((char_array_4[2] & 0x3c) >> 2);
+            char_array_3[2] = ((char_array_4[2] & 0x3) << 6) +   char_array_4[3];
+
+            for (i = 0; (i < 3); i++)
+            {
+                ret.push_back(char_array_3[i]);
+            }
+            i = 0;
+        }
+    }
+
+    if (i)
+    {
+        for (j = i; j <4; j++)
+        {
+            char_array_4[j] = 0;
+        }
+
+        for (j = 0; j <4; j++)
+        {
+            char_array_4[j] = base64_chars.find(char_array_4[j]);
+        }
+
+        char_array_3[0] = ((char_array_4[0]      ) << 2) + ((char_array_4[1] & 0x30) >> 4);
+        char_array_3[1] = ((char_array_4[1] & 0xf) << 4) + ((char_array_4[2] & 0x3c) >> 2);
+        char_array_3[2] = ((char_array_4[2] & 0x3) << 6) +   char_array_4[3];
+
+        for (j = 0; (j < i - 1); j++)
+        {
+            ret.push_back(char_array_3[j]);
+        }
+    }
+
+    return ret;
+}

backend/go/bark/gobark.cpp

@@ -48,7 +48,7 @@ int tts(char *text,int  threads, char *dst ) {
 
     // generate audio
     if (!bark_generate_audio(c, text, threads)) {
-        fprintf(stderr, "%s: An error occurred. If the problem persists, feel free to open an issue to report it.\n", __func__);
+        fprintf(stderr, "%s: An error occured. If the problem persists, feel free to open an issue to report it.\n", __func__);
         return 1;
     }
 

backend/go/bark/run.sh

@@ -1,3 +0,0 @@
-#!/bin/bash
-set -ex
-exec ./bark-cpp

backend/go/image/stablediffusion-ggml/Makefile

@@ -8,19 +8,12 @@ ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh
 # keep standard at C11 and C++11
 CXXFLAGS = -I. -I$(INCLUDE_PATH)/../../../../sources/stablediffusion-ggml.cpp/thirdparty -I$(INCLUDE_PATH)/../../../../sources/stablediffusion-ggml.cpp/ggml/include -I$(INCLUDE_PATH)/../../../../sources/stablediffusion-ggml.cpp -O3 -DNDEBUG -std=c++17 -fPIC
 
-GOCMD?=go
-CGO_LDFLAGS?=
-# Avoid parent make file overwriting CGO_LDFLAGS which is needed for hipblas
-CGO_LDFLAGS_SYCL=
-GO_TAGS?=
-LD_FLAGS?=
-
 # 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
 ifeq ($(BUILD_TYPE),cublas)
-	CMAKE_ARGS+=-DSD_CUDA=ON
+	CMAKE_ARGS+=-DGGML_CUDA=ON
 # If build type is openblas then we set -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
 # to CMAKE_ARGS automatically
 else ifeq ($(BUILD_TYPE),openblas)
@@ -28,50 +21,31 @@ else ifeq ($(BUILD_TYPE),openblas)
 # If build type is clblas (openCL) we set -DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
 else ifeq ($(BUILD_TYPE),clblas)
 	CMAKE_ARGS+=-DGGML_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++
+# If it's hipblas we do have also to set CC=/opt/rocm/llvm/bin/clang CXX=/opt/rocm/llvm/bin/clang++ 
 else ifeq ($(BUILD_TYPE),hipblas)
-	CMAKE_ARGS+=-DSD_HIPBLAS=ON
+	CMAKE_ARGS+=-DGGML_HIP=ON
 # If it's OSX, DO NOT embed the metal library - -DGGML_METAL_EMBED_LIBRARY=ON requires further investigation
 # But if it's OSX without metal, disable it here
 else ifeq ($(OS),Darwin)
 	ifneq ($(BUILD_TYPE),metal)
-		CMAKE_ARGS+=-DSD_METAL=OFF
+		CMAKE_ARGS+=-DGGML_METAL=OFF
 	else
-		CMAKE_ARGS+=-DSD_METAL=ON
+		CMAKE_ARGS+=-DGGML_METAL=ON
 		CMAKE_ARGS+=-DGGML_METAL_EMBED_LIBRARY=ON
 		TARGET+=--target ggml-metal
 	endif
 endif
 
-ifeq ($(BUILD_TYPE),sycl_f16)
-	CMAKE_ARGS+=-DGGML_SYCL=ON \
-		-DCMAKE_C_COMPILER=icx \
-		-DCMAKE_CXX_COMPILER=icpx \
-		-DSD_SYCL=ON \
-		-DGGML_SYCL_F16=ON
-	CC=icx
-	CXX=icpx
-	CGO_LDFLAGS_SYCL += -fsycl -L${DNNLROOT}/lib -ldnnl ${MKLROOT}/lib/intel64/libmkl_sycl.a -fiopenmp -fopenmp-targets=spir64 -lOpenCL
-	CGO_LDFLAGS_SYCL += $(shell pkg-config --libs mkl-static-lp64-gomp)
-	CGO_CXXFLAGS += -fiopenmp -fopenmp-targets=spir64
-	CGO_CXXFLAGS += $(shell pkg-config --cflags mkl-static-lp64-gomp )
-endif
+# ifeq ($(BUILD_TYPE),sycl_f16)
+# 	CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DGGML_SYCL_F16=ON -DSD_SYCL=ON -DGGML_SYCL_F16=ON
+# endif
 
-ifeq ($(BUILD_TYPE),sycl_f32)
-	CMAKE_ARGS+=-DGGML_SYCL=ON \
-		-DCMAKE_C_COMPILER=icx \
-		-DCMAKE_CXX_COMPILER=icpx \
-		-DSD_SYCL=ON
-	CC=icx
-	CXX=icpx
-	CGO_LDFLAGS_SYCL += -fsycl -L${DNNLROOT}/lib -ldnnl ${MKLROOT}/lib/intel64/libmkl_sycl.a -fiopenmp -fopenmp-targets=spir64 -lOpenCL
-	CGO_LDFLAGS_SYCL += $(shell pkg-config --libs mkl-static-lp64-gomp)
-	CGO_CXXFLAGS += -fiopenmp -fopenmp-targets=spir64
-	CGO_CXXFLAGS += $(shell pkg-config --cflags mkl-static-lp64-gomp )
-endif
+# ifeq ($(BUILD_TYPE),sycl_f32)
+# 	CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DSD_SYCL=ON
+# endif
 
 # warnings
-# CXXFLAGS += -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function
+CXXFLAGS += -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function
 
 # Find all .a archives in ARCHIVE_DIR
 # (ggml can have different backends cpu, cuda, etc., each backend generates a .a archive)
@@ -112,24 +86,11 @@ endif
 	$(MAKE) $(COMBINED_LIB)
 
 gosd.o:
-ifneq (,$(findstring sycl,$(BUILD_TYPE)))
-	+bash -c "source $(ONEAPI_VARS); \
-	$(CXX) $(CXXFLAGS) gosd.cpp -o gosd.o -c"
-else
 	$(CXX) $(CXXFLAGS) gosd.cpp -o gosd.o -c
-endif
 
 libsd.a: gosd.o
 	cp $(INCLUDE_PATH)/build/libstable-diffusion.a ./libsd.a
 	$(AR) rcs libsd.a gosd.o
 
-stablediffusion-ggml:
-	CGO_LDFLAGS="$(CGO_LDFLAGS) $(CGO_LDFLAGS_SYCL)" C_INCLUDE_PATH="$(INCLUDE_PATH)" LIBRARY_PATH="$(LIBRARY_PATH)" \
-	CC="$(CC)" CXX="$(CXX)" CGO_CXXFLAGS="$(CGO_CXXFLAGS)" \
-	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o ../../../../backend-assets/grpc/stablediffusion-ggml ./
-ifneq ($(UPX),)
-	$(UPX) ../../../../backend-assets/grpc/stablediffusion-ggml
-endif
-
 clean:
-	rm -rf gosd.o libsd.a build $(COMBINED_LIB)
+	rm -rf gosd.o libsd.a build $(COMBINED_LIB)

backend/go/llm/llama/llama.go

@@ -58,9 +58,6 @@ func (llm *LLM) Load(opts *pb.ModelOptions) error {
 	if opts.Embeddings {
 		llamaOpts = append(llamaOpts, llama.EnableEmbeddings)
 	}
-	if opts.Reranking {
-		llamaOpts = append(llamaOpts, llama.EnableReranking)
-	}
 	if opts.NGPULayers != 0 {
 		llamaOpts = append(llamaOpts, llama.SetGPULayers(int(opts.NGPULayers)))
 	}

backend/go/transcribe/whisper/whisper.go

@@ -74,7 +74,7 @@ func (sd *Whisper) AudioTranscription(opts *pb.TranscriptRequest) (pb.Transcript
 		context.SetTranslate(true)
 	}
 
-	if err := context.Process(data, nil, nil, nil); err != nil {
+	if err := context.Process(data, nil, nil); err != nil {
 		return pb.TranscriptResult{}, err
 	}
 

backend/go/vad/silero/vad.go

@@ -21,8 +21,8 @@ func (vad *VAD) Load(opts *pb.ModelOptions) error {
 		SampleRate: 16000,
 		//WindowSize:           1024,
 		Threshold:            0.5,
-		MinSilenceDurationMs: 100,
-		SpeechPadMs:          30,
+		MinSilenceDurationMs: 0,
+		SpeechPadMs:          0,
 	})
 	if err != nil {
 		return fmt.Errorf("create silero detector: %w", err)
@@ -35,10 +35,6 @@ func (vad *VAD) Load(opts *pb.ModelOptions) error {
 func (vad *VAD) VAD(req *pb.VADRequest) (pb.VADResponse, error) {
 	audio := req.Audio
 
-	if err := vad.detector.Reset(); err != nil {
-		return pb.VADResponse{}, fmt.Errorf("reset: %w", err)
-	}
-
 	segments, err := vad.detector.Detect(audio)
 	if err != nil {
 		return pb.VADResponse{}, fmt.Errorf("detect: %w", err)

backend/index.yaml

@@ -1,383 +0,0 @@
----
-## vLLM
-- &vllm
-  name: "cuda11-vllm"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-11-vllm"
-  license: apache-2.0
-  urls:
-    - https://github.com/vllm-project/vllm
-  tags:
-    - text-to-text
-    - multimodal
-    - GPTQ
-    - AWQ
-    - AutoRound
-    - INT4
-    - INT8
-    - FP8
-  icon: https://raw.githubusercontent.com/vllm-project/vllm/main/docs/assets/logos/vllm-logo-text-dark.png
-  description: |
-    vLLM is a fast and easy-to-use library for LLM inference and serving.
-    Originally developed in the Sky Computing Lab at UC Berkeley, vLLM has evolved into a community-driven project with contributions from both academia and industry.
-    vLLM is fast with:
-        State-of-the-art serving throughput
-        Efficient management of attention key and value memory with PagedAttention
-        Continuous batching of incoming requests
-        Fast model execution with CUDA/HIP graph
-        Quantizations: GPTQ, AWQ, AutoRound, INT4, INT8, and FP8
-        Optimized CUDA kernels, including integration with FlashAttention and FlashInfer
-        Speculative decoding
-        Chunked prefill
-  alias: "vllm"
-- !!merge <<: *vllm
-  name: "cuda12-vllm"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-vllm"
-- !!merge <<: *vllm
-  name: "rocm-vllm"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-rocm-hipblas-vllm"
-- !!merge <<: *vllm
-  name: "intel-sycl-f32-vllm"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-vllm"
-- !!merge <<: *vllm
-  name: "intel-sycl-f16-vllm"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-vllm"
-- !!merge <<: *vllm
-  name: "cuda11-vllm-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-vllm"
-- !!merge <<: *vllm
-  name: "cuda12-vllm-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-vllm"
-- !!merge <<: *vllm
-  name: "rocm-vllm-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-vllm"
-- !!merge <<: *vllm
-  name: "intel-sycl-f32-vllm-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-vllm"
-- !!merge <<: *vllm
-  name: "intel-sycl-f16-vllm-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-vllm"
-## Rerankers
-- name: "cuda11-rerankers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-11-rerankers"
-  alias: "cuda11-rerankers"
-- name: "cuda12-rerankers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-rerankers"
-  alias: "cuda12-rerankers"
-- name: "intel-sycl-f32-rerankers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-rerankers"
-  alias: "intel-sycl-f32-rerankers"
-- name: "intel-sycl-f16-rerankers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-rerankers"
-  alias: "intel-sycl-f16-rerankers"
-- name: "rocm-rerankers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-rocm-hipblas-rerankers"
-  alias: "rocm-rerankers"
-- name: "cuda11-rerankers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-rerankers"
-  alias: "rerankers"
-
-- name: "cuda12-rerankers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-rerankers"
-  alias: "rerankers"
-- name: "rocm-rerankers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-rerankers"
-  alias: "rerankers"
-
-- name: "intel-sycl-f32-rerankers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-rerankers"
-  alias: "rerankers"
-
-- name: "intel-sycl-f16-rerankers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-rerankers"
-  alias: "rerankers"
-## Transformers
-- &transformers
-  name: "cuda12-transformers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-transformers"
-  icon: https://camo.githubusercontent.com/26569a27b8a30a488dd345024b71dbc05da7ff1b2ba97bb6080c9f1ee0f26cc7/68747470733a2f2f68756767696e67666163652e636f2f64617461736574732f68756767696e67666163652f646f63756d656e746174696f6e2d696d616765732f7265736f6c76652f6d61696e2f7472616e73666f726d6572732f7472616e73666f726d6572735f61735f615f6d6f64656c5f646566696e6974696f6e2e706e67
-  alias: "transformers"
-  license: apache-2.0
-  description: |
-    Transformers acts as the model-definition framework for state-of-the-art machine learning models in text, computer vision, audio, video, and multimodal model, for both inference and training.
-    It centralizes the model definition so that this definition is agreed upon across the ecosystem. transformers is the pivot across frameworks: if a model definition is supported, it will be compatible with the majority of training frameworks (Axolotl, Unsloth, DeepSpeed, FSDP, PyTorch-Lightning, ...), inference engines (vLLM, SGLang, TGI, ...), and adjacent modeling libraries (llama.cpp, mlx, ...) which leverage the model definition from transformers.
-  urls:
-    - https://github.com/huggingface/transformers
-  tags:
-    - text-to-text
-    - multimodal
-- !!merge <<: *transformers
-  name: "rocm-transformers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-rocm-hipblas-transformers"
-- !!merge <<: *transformers
-  name: "intel-sycl-f32-transformers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-transformers"
-- !!merge <<: *transformers
-  name: "intel-sycl-f16-transformers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-transformers"
-- !!merge <<: *transformers
-  name: "cuda11-transformers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-transformers"
-- !!merge <<: *transformers
-  name: "cuda11-transformers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-11-transformers"
-- !!merge <<: *transformers
-  name: "cuda12-transformers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-transformers"
-- !!merge <<: *transformers
-  name: "rocm-transformers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-transformers"
-- !!merge <<: *transformers
-  name: "intel-sycl-f32-transformers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-transformers"
-- !!merge <<: *transformers
-  name: "intel-sycl-f16-transformers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-transformers"
-## Diffusers
-- &diffusers
-  icon: https://raw.githubusercontent.com/huggingface/diffusers/main/docs/source/en/imgs/diffusers_library.jpg
-  description: |
-    🤗 Diffusers is the go-to library for state-of-the-art pretrained diffusion models for generating images, audio, and even 3D structures of molecules. Whether you're looking for a simple inference solution or training your own diffusion models, 🤗 Diffusers is a modular toolbox that supports both.
-  urls:
-    - https://github.com/huggingface/diffusers
-  tags:
-    - image-generation
-    - video-generation
-    - diffusion-models
-  name: "cuda12-diffusers"
-  license: apache-2.0
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-diffusers"
-  alias: "diffusers"
-- !!merge <<: *diffusers
-  name: "rocm-diffusers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-rocm-hipblas-diffusers"
-- !!merge <<: *diffusers
-  name: "cuda11-diffusers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-11-diffusers"
-- !!merge <<: *diffusers
-  name: "intel-sycl-f32-diffusers"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-diffusers"
-- !!merge <<: *diffusers
-  name: "cuda11-diffusers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-diffusers"
-- !!merge <<: *diffusers
-  name: "cuda12-diffusers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-diffusers"
-- !!merge <<: *diffusers
-  name: "rocm-diffusers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-diffusers"
-- !!merge <<: *diffusers
-  name: "intel-sycl-f32-diffusers-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-diffusers"
-  ## exllama2
-- &exllama2
-  urls:
-    - https://github.com/turboderp-org/exllamav2
-  tags:
-    - text-to-text
-    - LLM
-    - EXL2
-  license: MIT
-  description: |
-    ExLlamaV2 is an inference library for running local LLMs on modern consumer GPUs.
-  name: "cuda11-exllama2"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-11-exllama2"
-  alias: "exllama2"
-- !!merge <<: *exllama2
-  name: "cuda12-exllama2"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-exllama2"
-- !!merge <<: *exllama2
-  name: "cuda11-exllama2-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-exllama2"
-- !!merge <<: *exllama2
-  name: "cuda12-exllama2-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-exllama2"
-## kokoro
-- &kokoro
-  icon: https://avatars.githubusercontent.com/u/166769057?v=4
-  description: |
-    Kokoro is an open-weight TTS model with 82 million parameters. Despite its lightweight architecture, it delivers comparable quality to larger models while being significantly faster and more cost-efficient. With Apache-licensed weights, Kokoro can be deployed anywhere from production environments to personal projects.
-  urls:
-    - https://huggingface.co/hexgrad/Kokoro-82M
-    - https://github.com/hexgrad/kokoro
-  tags:
-    - text-to-speech
-    - TTS
-    - LLM
-  license: apache-2.0
-  name: "cuda11-kokoro-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-kokoro"
-  alias: "kokoro"
-- !!merge <<: *kokoro
-  name: "cuda12-kokoro-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-kokoro"
-- !!merge <<: *kokoro
-  name: "rocm-kokoro-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-kokoro"
-- !!merge <<: *kokoro
-  name: "sycl-f32-kokoro"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-kokoro"
-- !!merge <<: *kokoro
-  name: "sycl-f16-kokoro"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-kokoro"
-- !!merge <<: *kokoro
-  name: "sycl-f16-kokoro-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-kokoro"
-- !!merge <<: *kokoro
-  name: "sycl-f32-kokoro-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-kokoro"
-## faster-whisper
-- &faster-whisper
-  icon: https://avatars.githubusercontent.com/u/1520500?s=200&v=4
-  description: |
-    faster-whisper is a reimplementation of OpenAI's Whisper model using CTranslate2, which is a fast inference engine for Transformer models.
-    This implementation is up to 4 times faster than openai/whisper for the same accuracy while using less memory. The efficiency can be further improved with 8-bit quantization on both CPU and GPU.
-  urls:
-    - https://github.com/SYSTRAN/faster-whisper
-  tags:
-    - speech-to-text
-    - Whisper
-  license: MIT
-  name: "cuda11-faster-whisper-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-faster-whisper"
-  alias: "faster-whisper"
-- !!merge <<: *faster-whisper
-  name: "cuda12-faster-whisper-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-faster-whisper"
-- !!merge <<: *faster-whisper
-  name: "rocm-faster-whisper-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-faster-whisper"
-- !!merge <<: *faster-whisper
-  name: "sycl-f32-faster-whisper"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-faster-whisper"
-- !!merge <<: *faster-whisper
-  name: "sycl-f16-faster-whisper"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-faster-whisper"
-- !!merge <<: *faster-whisper
-  name: "sycl-f32-faster-whisper-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-faster-whisper"
-- !!merge <<: *faster-whisper
-  name: "sycl-f16-faster-whisper-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-faster-whisper"
-## coqui
-- &coqui
-  urls:
-    - https://github.com/idiap/coqui-ai-TTS
-  description: |
-    🐸 Coqui TTS is a library for advanced Text-to-Speech generation.
-
-    🚀 Pretrained models in +1100 languages.
-
-    🛠️ Tools for training new models and fine-tuning existing models in any language.
-
-    📚 Utilities for dataset analysis and curation.
-  tags:
-    - text-to-speech
-    - TTS
-  license: mpl-2.0
-  name: "cuda11-coqui-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-coqui"
-  alias: "coqui"
-  icon: https://avatars.githubusercontent.com/u/1338804?s=200&v=4
-- !!merge <<: *coqui
-  name: "cuda12-coqui-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-coqui"
-- !!merge <<: *coqui
-  name: "rocm-coqui-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-coqui"
-- !!merge <<: *coqui
-  name: "sycl-f32-coqui"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-coqui"
-- !!merge <<: *coqui
-  name: "sycl-f16-coqui"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-coqui"
-- !!merge <<: *coqui
-  name: "sycl-f32-coqui-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-coqui"
-- !!merge <<: *coqui
-  name: "sycl-f16-coqui-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-coqui"
-## bark
-- &bark
-  urls:
-    - https://github.com/suno-ai/bark
-  description: |
-    Bark is a transformer-based text-to-audio model created by Suno. Bark can generate highly realistic, multilingual speech as well as other audio - including music, background noise and simple sound effects. The model can also produce nonverbal communications like laughing, sighing and crying. To support the research community, we are providing access to pretrained model checkpoints, which are ready for inference and available for commercial use.
-  tags:
-    - text-to-speech
-    - TTS
-  license: MIT
-  name: "cuda11-bark-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-bark"
-  alias: "bark"
-  icon: https://avatars.githubusercontent.com/u/99442120?s=200&v=4
-- !!merge <<: *bark
-  name: "cuda12-bark-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-bark"
-- !!merge <<: *bark
-  name: "rocm-bark-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-bark"
-- !!merge <<: *bark
-  name: "sycl-f32-bark"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f32-bark"
-- !!merge <<: *bark
-  name: "sycl-f16-bark"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-sycl-f16-bark"
-- !!merge <<: *bark
-  name: "sycl-f32-bark-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f32-bark"
-- !!merge <<: *bark
-  name: "sycl-f16-bark-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-sycl-f16-bark"
-- &barkcpp
-  urls:
-    - https://github.com/PABannier/bark.cpp
-  description: |
-      With bark.cpp, our goal is to bring real-time realistic multilingual text-to-speech generation to the community.
-
-      Plain C/C++ implementation without dependencies
-      AVX, AVX2 and AVX512 for x86 architectures
-      CPU and GPU compatible backends
-      Mixed F16 / F32 precision
-      4-bit, 5-bit and 8-bit integer quantization
-      Metal and CUDA backends
-
-      Models supported
-
-          Bark Small
-          Bark Large
-  tags:
-    - text-to-speech
-    - TTS
-  license: MIT
-  icon: https://github.com/PABannier/bark.cpp/raw/main/assets/banner.png
-  name: "bark-cpp"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-bark-cpp"
-  alias: "bark-cpp"
-- !!merge <<: *barkcpp
-  name: "bark-cpp-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-bark-cpp"
-  alias: "bark-cpp"
-## chatterbox
-- &chatterbox
-  urls:
-    - https://github.com/resemble-ai/chatterbox
-  description: |
-    Resemble AI's first production-grade open source TTS model. Licensed under MIT, Chatterbox has been benchmarked against leading closed-source systems like ElevenLabs, and is consistently preferred in side-by-side evaluations.
-    Whether you're working on memes, videos, games, or AI agents, Chatterbox brings your content to life. It's also the first open source TTS model to support emotion exaggeration control, a powerful feature that makes your voices stand out.
-  tags:
-    - text-to-speech
-    - TTS
-  license: MIT
-  icon: https://private-user-images.githubusercontent.com/660224/448166653-bd8c5f03-e91d-4ee5-b680-57355da204d1.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3NTAxOTE0MDAsIm5iZiI6MTc1MDE5MTEwMCwicGF0aCI6Ii82NjAyMjQvNDQ4MTY2NjUzLWJkOGM1ZjAzLWU5MWQtNGVlNS1iNjgwLTU3MzU1ZGEyMDRkMS5wbmc_WC1BbXotQWxnb3JpdGhtPUFXUzQtSE1BQy1TSEEyNTYmWC1BbXotQ3JlZGVudGlhbD1BS0lBVkNPRFlMU0E1M1BRSzRaQSUyRjIwMjUwNjE3JTJGdXMtZWFzdC0xJTJGczMlMkZhd3M0X3JlcXVlc3QmWC1BbXotRGF0ZT0yMDI1MDYxN1QyMDExNDBaJlgtQW16LUV4cGlyZXM9MzAwJlgtQW16LVNpZ25hdHVyZT1hMmI1NGY3OGFiZTlhNGFkNTVlYTY4NTIwMWEzODRiZGE4YzdhNGQ5MGNhNzE3MDYyYTA2NDIxYTkyYzhiODkwJlgtQW16LVNpZ25lZEhlYWRlcnM9aG9zdCJ9.mR9kM9xX0TdzPuSpuspCllHYQiq79dFQ2rtuNvjrl6w
-  name: "cuda11-chatterbox-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-11-chatterbox"
-  alias: "chatterbox"
-- !!merge <<: *chatterbox
-  name: "cuda12-chatterbox-development"
-  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-chatterbox"
-- !!merge <<: *chatterbox
-  name: "cuda11-chatterbox"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-11-chatterbox"
-- !!merge <<: *chatterbox
-  name: "cuda12-chatterbox"
-  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-chatterbox"

backend/python/autogptq/Makefile

@@ -0,0 +1,17 @@
+.PHONY: autogptq
+autogptq: protogen
+	bash install.sh
+
+.PHONY: protogen
+protogen: backend_pb2_grpc.py backend_pb2.py
+
+.PHONY: protogen-clean
+protogen-clean:
+	$(RM) backend_pb2_grpc.py backend_pb2.py
+
+backend_pb2_grpc.py backend_pb2.py:
+	python3 -m grpc_tools.protoc -I../.. --python_out=. --grpc_python_out=. backend.proto
+
+.PHONY: clean
+clean: protogen-clean
+	rm -rf venv __pycache__

backend/python/autogptq/README.md

@@ -0,0 +1,5 @@
+# Creating a separate environment for the autogptq project
+
+```
+make autogptq
+```

backend/python/autogptq/backend.py

@@ -0,0 +1,153 @@
+#!/usr/bin/env python3
+from concurrent import futures
+import argparse
+import signal
+import sys
+import os
+import time
+import base64
+
+import grpc
+import backend_pb2
+import backend_pb2_grpc
+
+from auto_gptq import AutoGPTQForCausalLM
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from transformers import TextGenerationPipeline
+
+_ONE_DAY_IN_SECONDS = 60 * 60 * 24
+
+# 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'))
+
+# 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:
+            device = "cuda:0"
+            if request.Device != "":
+                device = request.Device
+
+            # support loading local model files
+            model_path = os.path.join(os.environ.get('MODELS_PATH', './'), request.Model)
+            tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=True, trust_remote_code=request.TrustRemoteCode)
+
+            # support model `Qwen/Qwen-VL-Chat-Int4`
+            if "qwen-vl" in request.Model.lower():
+                self.model_name = "Qwen-VL-Chat"
+                model = AutoModelForCausalLM.from_pretrained(model_path, 
+                    trust_remote_code=request.TrustRemoteCode,
+                    device_map="auto").eval()
+            else:
+                model = AutoGPTQForCausalLM.from_quantized(model_path,
+                    model_basename=request.ModelBaseName,
+                    use_safetensors=True,
+                    trust_remote_code=request.TrustRemoteCode,
+                    device=device,
+                    use_triton=request.UseTriton,
+                    quantize_config=None)
+            
+            self.model = model
+            self.tokenizer = tokenizer
+        except Exception as err:
+            return backend_pb2.Result(success=False, message=f"Unexpected {err=}, {type(err)=}")
+        return backend_pb2.Result(message="Model loaded successfully", success=True)
+
+    def Predict(self, request, context):
+        penalty = 1.0
+        if request.Penalty != 0.0:
+            penalty = request.Penalty
+        tokens = 512
+        if request.Tokens != 0:
+            tokens = request.Tokens
+        top_p = 0.95
+        if request.TopP != 0.0:
+            top_p = request.TopP
+
+        
+        prompt_images = self.recompile_vl_prompt(request)
+        compiled_prompt = prompt_images[0]
+        print(f"Prompt: {compiled_prompt}", file=sys.stderr)
+
+        # Implement Predict RPC
+        pipeline = TextGenerationPipeline(
+            model=self.model, 
+            tokenizer=self.tokenizer,
+            max_new_tokens=tokens,
+            temperature=request.Temperature,
+            top_p=top_p,
+            repetition_penalty=penalty,
+            )
+        t = pipeline(compiled_prompt)[0]["generated_text"]
+        print(f"generated_text: {t}", file=sys.stderr)
+        
+        if compiled_prompt in t:
+            t = t.replace(compiled_prompt, "")
+        # house keeping. Remove the image files from /tmp folder
+        for img_path in prompt_images[1]:
+            try:
+                os.remove(img_path)
+            except Exception as e:
+                print(f"Error removing image file: {img_path}, {e}", file=sys.stderr)
+
+        return backend_pb2.Result(message=bytes(t, encoding='utf-8'))
+
+    def PredictStream(self, request, context):
+        # Implement PredictStream RPC
+        #for reply in some_data_generator():
+        #    yield reply
+        # Not implemented yet
+        return self.Predict(request, context)
+
+    def recompile_vl_prompt(self, request):
+        prompt = request.Prompt
+        image_paths = []
+
+        if "qwen-vl" in self.model_name.lower():
+            # request.Images is an array which contains base64 encoded images. Iterate the request.Images array, decode and save each image to /tmp folder with a random filename.
+            # Then, save the image file paths to an array "image_paths".
+            # read "request.Prompt", replace "[img-%d]" with the image file paths in the order they appear in "image_paths". Save the new prompt to "prompt".
+            for i, img in enumerate(request.Images):
+                timestamp = str(int(time.time() * 1000))  # Generate timestamp
+                img_path = f"/tmp/vl-{timestamp}.jpg"  # Use timestamp in filename
+                with open(img_path, "wb") as f:
+                    f.write(base64.b64decode(img))
+                image_paths.append(img_path)
+                prompt = prompt.replace(f"[img-{i}]", "<img>" + img_path + "</img>,")
+        else:
+            prompt = request.Prompt
+        return (prompt, image_paths)
+
+def serve(address):
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
+    backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
+    server.add_insecure_port(address)
+    server.start()
+    print("Server started. Listening on: " + address, file=sys.stderr)
+
+    # Define the signal handler function
+    def signal_handler(sig, frame):
+        print("Received termination signal. Shutting down...")
+        server.stop(0)
+        sys.exit(0)
+
+    # Set the signal handlers for SIGINT and SIGTERM
+    signal.signal(signal.SIGINT, signal_handler)
+    signal.signal(signal.SIGTERM, signal_handler)
+
+    try:
+        while True:
+            time.sleep(_ONE_DAY_IN_SECONDS)
+    except KeyboardInterrupt:
+        server.stop(0)
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Run the gRPC server.")
+    parser.add_argument(
+        "--addr", default="localhost:50051", help="The address to bind the server to."
+    )
+    args = parser.parse_args()
+
+    serve(args.addr)

backend/python/autogptq/install.sh

@@ -1,12 +1,7 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 # This is here because the Intel pip index is broken and returns 200 status codes for every package name, it just doesn't return any package links.
 # This makes uv think that the package exists in the Intel pip index, and by default it stops looking at other pip indexes once it finds a match.

backend/python/autogptq/requirements-cublas11.txt

@@ -0,0 +1,2 @@
+--extra-index-url https://download.pytorch.org/whl/cu118
+torch==2.4.1+cu118

backend/python/autogptq/requirements-cublas12.txt

@@ -0,0 +1 @@
+torch==2.4.1

backend/python/autogptq/requirements-hipblas.txt

@@ -0,0 +1,2 @@
+--extra-index-url https://download.pytorch.org/whl/rocm6.0
+torch==2.4.1+rocm6.0

backend/python/autogptq/requirements-intel.txt

@@ -1,11 +1,6 @@
 --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
 intel-extension-for-pytorch==2.3.110+xpu
 torch==2.3.1+cxx11.abi
-torchaudio==2.3.1+cxx11.abi
-transformers==4.46.3
-chatterbox-tts
-accelerate
 oneccl_bind_pt==2.3.100+xpu
 optimum[openvino]
-setuptools
-accelerate
+setuptools

backend/python/autogptq/requirements.txt

@@ -0,0 +1,6 @@
+accelerate
+auto-gptq==0.7.1
+grpcio==1.71.0
+protobuf
+certifi
+transformers

backend/python/autogptq/run.sh

@@ -0,0 +1,4 @@
+#!/bin/bash
+source $(dirname $0)/../common/libbackend.sh
+
+startBackend $@

backend/python/autogptq/test.sh

@@ -0,0 +1,6 @@
+#!/bin/bash
+set -e
+
+source $(dirname $0)/../common/libbackend.sh
+
+runUnittests

backend/python/bark/Makefile

@@ -22,7 +22,7 @@ protogen-clean:
 	$(RM) backend_pb2_grpc.py backend_pb2.py
 
 backend_pb2_grpc.py backend_pb2.py:
-	python3 -m grpc_tools.protoc -I../.. -I./ --python_out=. --grpc_python_out=. backend.proto
+	python3 -m grpc_tools.protoc -I../.. --python_out=. --grpc_python_out=. backend.proto
 
 .PHONY: clean
 clean: protogen-clean

backend/python/bark/backend.py

@@ -61,12 +61,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         return backend_pb2.Result(success=True)
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS),
-        options=[
-            ('grpc.max_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_send_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_receive_message_length', 50 * 1024 * 1024),  # 50MB
-        ])
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()

backend/python/bark/install.sh

@@ -1,12 +1,7 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 # This is here because the Intel pip index is broken and returns 200 status codes for every package name, it just doesn't return any package links.
 # This makes uv think that the package exists in the Intel pip index, and by default it stops looking at other pip indexes once it finds a match.

backend/python/bark/run.sh

@@ -1,9 +1,4 @@
 #!/bin/bash
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 startBackend $@

backend/python/bark/test.sh

@@ -1,11 +1,6 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 runUnittests

backend/python/chatterbox/Makefile

@@ -1,29 +0,0 @@
-.PHONY: coqui
-coqui: protogen
-	bash install.sh
-
-.PHONY: run
-run: protogen
-	@echo "Running coqui..."
-	bash run.sh
-	@echo "coqui run."
-
-.PHONY: test
-test: protogen
-	@echo "Testing coqui..."
-	bash test.sh
-	@echo "coqui tested."
-
-.PHONY: protogen
-protogen: backend_pb2_grpc.py backend_pb2.py
-
-.PHONY: protogen-clean
-protogen-clean:
-	$(RM) backend_pb2_grpc.py backend_pb2.py
-
-backend_pb2_grpc.py backend_pb2.py:
-	python3 -m grpc_tools.protoc -I../.. -I./ --python_out=. --grpc_python_out=. backend.proto
-
-.PHONY: clean
-clean: protogen-clean
-	rm -rf venv __pycache__

backend/python/chatterbox/backend.py

@@ -1,117 +0,0 @@
-#!/usr/bin/env python3
-"""
-This is an extra gRPC server of LocalAI for Bark TTS
-"""
-from concurrent import futures
-import time
-import argparse
-import signal
-import sys
-import os
-import backend_pb2
-import backend_pb2_grpc
-
-import torch
-import torchaudio as ta
-from chatterbox.tts import ChatterboxTTS
-
-import grpc
-
-
-_ONE_DAY_IN_SECONDS = 60 * 60 * 24
-
-# 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'))
-COQUI_LANGUAGE = os.environ.get('COQUI_LANGUAGE', None)
-
-# Implement the BackendServicer class with the service methods
-class BackendServicer(backend_pb2_grpc.BackendServicer):
-    """
-    BackendServicer is the class that implements the gRPC service
-    """
-    def Health(self, request, context):
-        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
-    def LoadModel(self, request, context):
-
-        # Get device
-        # device = "cuda" if request.CUDA else "cpu"
-        if torch.cuda.is_available():
-            print("CUDA is available", file=sys.stderr)
-            device = "cuda"
-        else:
-            print("CUDA is not available", file=sys.stderr)
-            device = "cpu"
-
-        if not torch.cuda.is_available() and request.CUDA:
-            return backend_pb2.Result(success=False, message="CUDA is not available")
-
-        self.AudioPath = None
-
-        if os.path.isabs(request.AudioPath):
-            self.AudioPath = request.AudioPath
-        elif request.AudioPath and request.ModelFile != "" and not os.path.isabs(request.AudioPath):
-            # get base path of modelFile
-            modelFileBase = os.path.dirname(request.ModelFile)
-            # modify LoraAdapter to be relative to modelFileBase
-            self.AudioPath = os.path.join(modelFileBase, request.AudioPath)
-
-        try:
-            print("Preparing models, please wait", file=sys.stderr)
-            self.model = ChatterboxTTS.from_pretrained(device=device)
-        except Exception as err:
-            return backend_pb2.Result(success=False, message=f"Unexpected {err=}, {type(err)=}")
-        # Implement your logic here for the LoadModel service
-        # Replace this with your desired response
-        return backend_pb2.Result(message="Model loaded successfully", success=True)
-
-    def TTS(self, request, context):
-        try:
-            # Generate audio using ChatterboxTTS
-            if self.AudioPath is not None:
-                wav = self.model.generate(request.text, audio_prompt_path=self.AudioPath)
-            else:
-                wav = self.model.generate(request.text)
-            
-            # Save the generated audio
-            ta.save(request.dst, wav, self.model.sr)
-            
-        except Exception as err:
-            return backend_pb2.Result(success=False, message=f"Unexpected {err=}, {type(err)=}")
-        return backend_pb2.Result(success=True)
-
-def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS),
-        options=[
-            ('grpc.max_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_send_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_receive_message_length', 50 * 1024 * 1024),  # 50MB
-        ])
-    backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
-    server.add_insecure_port(address)
-    server.start()
-    print("Server started. Listening on: " + address, file=sys.stderr)
-
-    # Define the signal handler function
-    def signal_handler(sig, frame):
-        print("Received termination signal. Shutting down...")
-        server.stop(0)
-        sys.exit(0)
-
-    # Set the signal handlers for SIGINT and SIGTERM
-    signal.signal(signal.SIGINT, signal_handler)
-    signal.signal(signal.SIGTERM, signal_handler)
-
-    try:
-        while True:
-            time.sleep(_ONE_DAY_IN_SECONDS)
-    except KeyboardInterrupt:
-        server.stop(0)
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description="Run the gRPC server.")
-    parser.add_argument(
-        "--addr", default="localhost:50051", help="The address to bind the server to."
-    )
-    args = parser.parse_args()
-
-    serve(args.addr)

backend/python/chatterbox/requirements-cpu.txt

@@ -1,5 +0,0 @@
-accelerate
-torch==2.6.0
-torchaudio==2.6.0
-transformers==4.46.3
-chatterbox-tts

backend/python/chatterbox/requirements-cublas11.txt

@@ -1,6 +0,0 @@
---extra-index-url https://download.pytorch.org/whl/cu118
-torch==2.6.0+cu118
-torchaudio==2.6.0+cu118
-transformers==4.46.3
-chatterbox-tts
-accelerate

backend/python/chatterbox/requirements-cublas12.txt

@@ -1,5 +0,0 @@
-torch==2.6.0
-torchaudio==2.6.0
-transformers==4.46.3
-chatterbox-tts
-accelerate

backend/python/chatterbox/requirements-hipblas.txt

@@ -1,6 +0,0 @@
---extra-index-url https://download.pytorch.org/whl/rocm6.0
-torch==2.6.0+rocm6.1
-torchaudio==2.6.0+rocm6.1
-transformers==4.46.3
-chatterbox-tts
-accelerate

backend/python/chatterbox/requirements.txt

@@ -1,5 +0,0 @@
-grpcio==1.71.0
-protobuf
-certifi
-packaging
-setuptools

backend/python/chatterbox/run.sh

@@ -1,9 +0,0 @@
-#!/bin/bash
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
-
-startBackend $@

backend/python/chatterbox/test.py

@@ -1,82 +0,0 @@
-"""
-A test script to test the gRPC service
-"""
-import unittest
-import subprocess
-import time
-import backend_pb2
-import backend_pb2_grpc
-
-import grpc
-
-
-class TestBackendServicer(unittest.TestCase):
-    """
-    TestBackendServicer is the class that tests the gRPC service
-    """
-    def setUp(self):
-        """
-        This method sets up the gRPC service by starting the server
-        """
-        self.service = subprocess.Popen(["python3", "backend.py", "--addr", "localhost:50051"])
-        time.sleep(30)
-
-    def tearDown(self) -> None:
-        """
-        This method tears down the gRPC service by terminating the server
-        """
-        self.service.terminate()
-        self.service.wait()
-
-    def test_server_startup(self):
-        """
-        This method tests if the server starts up successfully
-        """
-        try:
-            self.setUp()
-            with grpc.insecure_channel("localhost:50051") as channel:
-                stub = backend_pb2_grpc.BackendStub(channel)
-                response = stub.Health(backend_pb2.HealthMessage())
-                self.assertEqual(response.message, b'OK')
-        except Exception as err:
-            print(err)
-            self.fail("Server failed to start")
-        finally:
-            self.tearDown()
-
-    def test_load_model(self):
-        """
-        This method tests if the model is loaded successfully
-        """
-        try:
-            self.setUp()
-            with grpc.insecure_channel("localhost:50051") as channel:
-                stub = backend_pb2_grpc.BackendStub(channel)
-                response = stub.LoadModel(backend_pb2.ModelOptions())
-                print(response)
-                self.assertTrue(response.success)
-                self.assertEqual(response.message, "Model loaded successfully")
-        except Exception as err:
-            print(err)
-            self.fail("LoadModel service failed")
-        finally:
-            self.tearDown()
-
-    def test_tts(self):
-        """
-        This method tests if the embeddings are generated successfully
-        """
-        try:
-            self.setUp()
-            with grpc.insecure_channel("localhost:50051") as channel:
-                stub = backend_pb2_grpc.BackendStub(channel)
-                response = stub.LoadModel(backend_pb2.ModelOptions())
-                self.assertTrue(response.success)
-                tts_request = backend_pb2.TTSRequest(text="80s TV news production music hit for tonight's biggest story")
-                tts_response = stub.TTS(tts_request)
-                self.assertIsNotNone(tts_response)
-        except Exception as err:
-            print(err)
-            self.fail("TTS service failed")
-        finally:
-            self.tearDown()

backend/python/chatterbox/test.sh

@@ -1,11 +0,0 @@
-#!/bin/bash
-set -e
-
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
-
-runUnittests

backend/python/common/libbackend.sh

@@ -176,13 +176,13 @@ function startBackend() {
     ensureVenv
 
     if [ ! -z ${BACKEND_FILE} ]; then
-        exec ${EDIR}/venv/bin/python ${BACKEND_FILE} $@
+        exec python ${BACKEND_FILE} $@
     elif [ -e "${MY_DIR}/server.py" ]; then
-        exec ${EDIR}/venv/bin/python ${MY_DIR}/server.py $@
+        exec python ${MY_DIR}/server.py $@
     elif [ -e "${MY_DIR}/backend.py" ]; then
-        exec ${EDIR}/venv/bin/python ${MY_DIR}/backend.py $@
+        exec python ${MY_DIR}/backend.py $@
     elif [ -e "${MY_DIR}/${BACKEND_NAME}.py" ]; then
-        exec ${EDIR}/venv/bin/python ${MY_DIR}/${BACKEND_NAME}.py $@
+        exec python ${MY_DIR}/${BACKEND_NAME}.py $@
     fi
 }
 

backend/python/common/template/install.sh

@@ -1,12 +1,7 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 # This is here because the Intel pip index is broken and returns 200 status codes for every package name, it just doesn't return any package links.
 # This makes uv think that the package exists in the Intel pip index, and by default it stops looking at other pip indexes once it finds a match.

backend/python/common/template/protogen.sh

@@ -1,11 +1,6 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
-python3 -m grpc_tools.protoc -I../.. -I./ --python_out=. --grpc_python_out=. backend.proto
+python3 -m grpc_tools.protoc -I../.. --python_out=. --grpc_python_out=. backend.proto

backend/python/common/template/run.sh

@@ -1,9 +1,4 @@
 #!/bin/bash
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 startBackend $@

backend/python/common/template/test.sh

@@ -1,11 +1,6 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 runUnittests

backend/python/coqui/Makefile

@@ -22,7 +22,7 @@ protogen-clean:
 	$(RM) backend_pb2_grpc.py backend_pb2.py
 
 backend_pb2_grpc.py backend_pb2.py:
-	python3 -m grpc_tools.protoc -I../.. -I./ --python_out=. --grpc_python_out=. backend.proto
+	python3 -m grpc_tools.protoc -I../.. --python_out=. --grpc_python_out=. backend.proto
 
 .PHONY: clean
 clean: protogen-clean

backend/python/coqui/backend.py

@@ -66,7 +66,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
     def TTS(self, request, context):
         try:
-            # if model is multilingual add language from request or env as fallback
+            # if model is multilangual add language from request or env as fallback
             lang = request.language or COQUI_LANGUAGE
             if lang == "":
                 lang = None
@@ -86,12 +86,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         return backend_pb2.Result(success=True)
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS),
-        options=[
-            ('grpc.max_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_send_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_receive_message_length', 50 * 1024 * 1024),  # 50MB
-        ])
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()

backend/python/coqui/install.sh

@@ -1,12 +1,7 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 # This is here because the Intel pip index is broken and returns 200 status codes for every package name, it just doesn't return any package links.
 # This makes uv think that the package exists in the Intel pip index, and by default it stops looking at other pip indexes once it finds a match.

backend/python/coqui/run.sh

@@ -1,9 +1,4 @@
 #!/bin/bash
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 startBackend $@

backend/python/coqui/test.sh

@@ -1,11 +1,6 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 runUnittests

backend/python/diffusers/Makefile

@@ -32,7 +32,7 @@ protogen-clean:
 	$(RM) backend_pb2_grpc.py backend_pb2.py
 
 backend_pb2_grpc.py backend_pb2.py:
-	python3 -m grpc_tools.protoc -I../.. -I./ --python_out=. --grpc_python_out=. backend.proto
+	python3 -m grpc_tools.protoc -I../.. --python_out=. --grpc_python_out=. backend.proto
 
 .PHONY: clean
 clean: protogen-clean

backend/python/diffusers/backend.py

@@ -19,7 +19,7 @@ import grpc
 
 from diffusers import SanaPipeline, StableDiffusion3Pipeline, StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, \
     EulerAncestralDiscreteScheduler, FluxPipeline, FluxTransformer2DModel
-from diffusers import StableDiffusionImg2ImgPipeline, AutoPipelineForText2Image, ControlNetModel, StableVideoDiffusionPipeline, Lumina2Text2ImgPipeline
+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 compel import Compel, ReturnedEmbeddingsType
@@ -168,13 +168,9 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             # We are storing all the options in a dict so we can use it later when
             # generating the images
             for opt in options:
-                if ":" not in opt:
-                    continue
                 key, value = opt.split(":")
                 self.options[key] = value
 
-            print(f"Options: {self.options}", file=sys.stderr)
-
             local = False
             modelFile = request.Model
 
@@ -291,12 +287,6 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
                     if request.LowVRAM:
                         self.pipe.enable_model_cpu_offload()
-            elif request.PipelineType == "Lumina2Text2ImgPipeline":
-                self.pipe = Lumina2Text2ImgPipeline.from_pretrained(
-                    request.Model,
-                    torch_dtype=torch.bfloat16)
-                if request.LowVRAM:
-                    self.pipe.enable_model_cpu_offload()
             elif request.PipelineType == "SanaPipeline":
                 self.pipe = SanaPipeline.from_pretrained(
                     request.Model,
@@ -526,12 +516,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS),
-        options=[
-            ('grpc.max_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_send_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_receive_message_length', 50 * 1024 * 1024),  # 50MB
-        ])
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()

backend/python/diffusers/install.sh

@@ -1,12 +1,7 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 # This is here because the Intel pip index is broken and returns 200 status codes for every package name, it just doesn't return any package links.
 # This makes uv think that the package exists in the Intel pip index, and by default it stops looking at other pip indexes once it finds a match.

backend/python/diffusers/run.sh

@@ -1,9 +1,4 @@
 #!/bin/bash
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 startBackend $@

backend/python/diffusers/test.sh

@@ -1,11 +1,6 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 runUnittests

backend/python/exllama2/Makefile

@@ -16,7 +16,7 @@ protogen-clean:
 	$(RM) backend_pb2_grpc.py backend_pb2.py
 
 backend_pb2_grpc.py backend_pb2.py:
-	python3 -m grpc_tools.protoc -I../.. -I./ --python_out=. --grpc_python_out=. backend.proto
+	python3 -m grpc_tools.protoc -I../.. --python_out=. --grpc_python_out=. backend.proto
 
 .PHONY: clean
 clean: protogen-clean

backend/python/exllama2/backend.py

@@ -105,12 +105,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS),
-        options=[
-            ('grpc.max_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_send_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_receive_message_length', 50 * 1024 * 1024),  # 50MB
-        ])
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()

backend/python/exllama2/install.sh

@@ -5,12 +5,7 @@ LIMIT_TARGETS="cublas"
 EXTRA_PIP_INSTALL_FLAGS="--no-build-isolation"
 EXLLAMA2_VERSION=c0ddebaaaf8ffd1b3529c2bb654e650bce2f790f
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 installRequirements
 

backend/python/exllama2/run.sh

@@ -1,11 +1,6 @@
 #!/bin/bash
 LIMIT_TARGETS="cublas"
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 startBackend $@

backend/python/exllama2/test.sh

@@ -1,11 +1,6 @@
 #!/bin/bash
 set -e
 
-backend_dir=$(dirname $0)
-if [ -d $backend_dir/common ]; then
-    source $backend_dir/common/libbackend.sh
-else
-    source $backend_dir/../common/libbackend.sh
-fi
+source $(dirname $0)/../common/libbackend.sh
 
 runUnittests

backend/python/faster-whisper/backend.py

@@ -62,12 +62,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         return backend_pb2.TranscriptResult(segments=resultSegments, text=text)
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS),
-        options=[
-            ('grpc.max_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_send_message_length', 50 * 1024 * 1024),  # 50MB
-            ('grpc.max_receive_message_length', 50 * 1024 * 1024),  # 50MB
-        ])
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()