Merge branch 'master' into fix_aarch64

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

.github/ci/modelslist.go

@@ -75,7 +75,7 @@ var modelPageTemplate string = `
     <div class="container mx-auto px-4 py-4">
         <div class="flex items-center justify-between">
             <div class="flex items-center">
-                <a href="/" class="text-white text-xl font-bold"><img src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo" class="h-10 mr-3 border-2 border-gray-300 shadow rounded"></a>
+                <a href="/" class="text-white text-xl font-bold"><img src="https://github.com/mudler/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo" class="h-10 mr-3 border-2 border-gray-300 shadow rounded"></a>
                 <a href="/" class="text-white text-xl font-bold">LocalAI</a>
             </div>
             <!-- Menu button for small screens -->
@@ -92,9 +92,9 @@ var modelPageTemplate string = `
         <!-- Collapsible menu for small screens -->
         <div class="hidden lg:hidden" id="mobile-menu">
             <div class="pt-4 pb-3 border-t border-gray-700">
-                
+
                 <a href="https://localai.io" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1" target="_blank" ><i class="fas fa-book-reader pr-2"></i> Documentation</a>
-               
+
             </div>
         </div>
     </div>
@@ -114,17 +114,17 @@ var modelPageTemplate string = `
 
 	<h2 class="text-center text-3xl font-semibold text-gray-100">
 
-	 🖼️ Available {{.AvailableModels}} models</i> repositories     <a href="https://localai.io/models/" target="_blank" >
+	 🖼️ Available {{.AvailableModels}} models</i> <a href="https://localai.io/models/" target="_blank" >
 			<i class="fas fa-circle-info pr-2"></i>
-		</a></h2> 
+		</a></h2>
 
-	<h3>	  
-	Refer to <a href="https://localai.io/models" target=_blank> Model gallery</a> for more information on how to use the models with LocalAI.
+	<h3>
+	Refer to the Model gallery <a href="https://localai.io/models/" target="_blank" ><i class="fas fa-circle-info pr-2"></i></a> for more information on how to use the models with LocalAI.<br>
 
 	You can install models with the CLI command <code>local-ai models install <model-name></code>. or by using the WebUI.
 	</h3>
-  
-	<input class="form-control appearance-none block w-full mt-5 px-3 py-2 text-base font-normal text-gray-300 pb-2 mb-5 bg-gray-800 bg-clip-padding border border-solid border-gray-600 rounded transition ease-in-out m-0 focus:text-gray-300 focus:bg-gray-900 focus:border-blue-500 focus:outline-none" type="search" 
+
+	<input class="form-control appearance-none block w-full mt-5 px-3 py-2 text-base font-normal text-gray-300 pb-2 mb-5 bg-gray-800 bg-clip-padding border border-solid border-gray-600 rounded transition ease-in-out m-0 focus:text-gray-300 focus:bg-gray-900 focus:border-blue-500 focus:outline-none" type="search"
 	id="searchbox" placeholder="Live search keyword..">
 	  <div class="dark grid grid-cols-1 grid-rows-1 md:grid-cols-3 block rounded-lg shadow-secondary-1 dark:bg-surface-dark">
 		{{ range $_, $model := .Models }}
@@ -139,10 +139,10 @@ var modelPageTemplate string = `
 			</div>
 	  		<div class="p-6 text-surface dark:text-white">
 				<h5 class="mb-2 text-xl font-medium leading-tight">{{$model.Name}}</h5>
-				
-				   
+
+
 				<p class="mb-4 text-base truncate">{{ $model.Description }}</p>
-		
+
 			</div>
 			<div class="px-6 pt-4 pb-2">
 
@@ -178,7 +178,7 @@ var modelPageTemplate string = `
                     {{ $model.Description }}
 
                     </p>
-                    
+
                     <p class="text-base leading-relaxed text-gray-500 dark:text-gray-400">
                     To install the model with the CLI, run: <br>
                     <code> local-ai models install {{$model.Name}} </code> <br>
@@ -193,7 +193,7 @@ var modelPageTemplate string = `
                     <ul>
                     {{ range $_, $u := $model.URLs }}
                     <li><a href="{{ $u }}" target=_blank><i class="fa-solid fa-link"></i> {{ $u }}</a></li>
-                    {{ end }}  
+                    {{ end }}
                     </ul>
                     </p>
                 </div>
@@ -209,7 +209,7 @@ var modelPageTemplate string = `
 			</div>
 		</div>
 		</div>
-		{{ end }}      
+		{{ end }}
 
 		</div>
   </div>
@@ -221,10 +221,10 @@ var lazyLoadInstance = new LazyLoad({
 });
 
 let cards = document.querySelectorAll('.box')
-    
+
 function liveSearch() {
     let search_query = document.getElementById("searchbox").value;
-    
+
     //Use innerText if all contents are visible
     //Use textContent for including hidden elements
     for (var i = 0; i < cards.length; i++) {
@@ -238,8 +238,8 @@ function liveSearch() {
 }
 
 //A little delay
-let typingTimer;               
-let typeInterval = 500;  
+let typingTimer;
+let typeInterval = 500;
 let searchInput = document.getElementById('searchbox');
 
 searchInput.addEventListener('keyup', () => {

.github/release.yml

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

.github/workflows/image-pr.yml

@@ -32,7 +32,7 @@ jobs:
     strategy:
       # Pushing with all jobs in parallel
       # eats the bandwidth of all the nodes
-      max-parallel: ${{ github.event_name != 'pull_request' && 2 || 4 }}
+      max-parallel: ${{ github.event_name != 'pull_request' && 4 || 8 }}
       matrix:
         include:
           - build-type: ''
@@ -46,7 +46,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "1"
+            cuda-minor-version: "4"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-ffmpeg'
@@ -119,7 +119,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "1"
+            cuda-minor-version: "4"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-ffmpeg-core'
@@ -127,4 +127,13 @@ jobs:
             image-type: 'core'
             runs-on: 'ubuntu-latest'
             base-image: "ubuntu:22.04"
-            makeflags: "--jobs=4 --output-sync=target"
+            makeflags: "--jobs=4 --output-sync=target"
+          - build-type: 'vulkan'
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-vulkan-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:22.04"
+            makeflags: "--jobs=4 --output-sync=target"

.github/workflows/image.yml

@@ -39,7 +39,7 @@ jobs:
     strategy:
       # Pushing with all jobs in parallel
       # eats the bandwidth of all the nodes
-      max-parallel: ${{ github.event_name != 'pull_request' && 2 || 4 }}
+      max-parallel: ${{ github.event_name != 'pull_request' && 6 || 10 }}
       matrix:
         include:
           # Extra images
@@ -75,7 +75,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "1"
+            cuda-minor-version: "4"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12'
@@ -100,7 +100,7 @@ jobs:
             makeflags: "--jobs=3 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "1"
+            cuda-minor-version: "4"
             platforms: 'linux/amd64'
             tag-latest: 'auto'
             tag-suffix: '-cublas-cuda12-ffmpeg'
@@ -232,7 +232,7 @@ jobs:
             grpc-base-image: "ubuntu:22.04"
             runs-on: 'arc-runner-set'
             makeflags: "--jobs=3 --output-sync=target"
-  
+
   core-image-build:
     uses: ./.github/workflows/image_build.yml
     with:
@@ -257,6 +257,7 @@ jobs:
       quayUsername: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
       quayPassword: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
     strategy:
+      max-parallel: ${{ github.event_name != 'pull_request' && 2 || 4 }}
       matrix:
         include:
           - build-type: ''
@@ -266,7 +267,7 @@ jobs:
             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'
@@ -280,18 +281,18 @@ jobs:
             ffmpeg: ''
             image-type: 'core'
             base-image: "ubuntu:22.04"
-            runs-on: 'ubuntu-latest'
+            runs-on: 'arc-runner-set'
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "1"
+            cuda-minor-version: "4"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-core'
             ffmpeg: ''
             image-type: 'core'
             base-image: "ubuntu:22.04"
-            runs-on: 'ubuntu-latest'
+            runs-on: 'arc-runner-set'
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "11"
@@ -301,17 +302,27 @@ jobs:
             tag-suffix: '-cublas-cuda11-ffmpeg-core'
             ffmpeg: 'true'
             image-type: 'core'
-            runs-on: 'ubuntu-latest'
+            runs-on: 'arc-runner-set'
             base-image: "ubuntu:22.04"
             makeflags: "--jobs=4 --output-sync=target"
           - build-type: 'cublas'
             cuda-major-version: "12"
-            cuda-minor-version: "1"
+            cuda-minor-version: "4"
             platforms: 'linux/amd64'
             tag-latest: 'false'
             tag-suffix: '-cublas-cuda12-ffmpeg-core'
             ffmpeg: 'true'
             image-type: 'core'
-            runs-on: 'ubuntu-latest'
+            runs-on: 'arc-runner-set'
+            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'
+            latest-image: 'latest-vulkan-ffmpeg-core'
+            ffmpeg: 'true'
+            image-type: 'core'
+            runs-on: 'arc-runner-set'
             base-image: "ubuntu:22.04"
             makeflags: "--jobs=4 --output-sync=target"

.github/workflows/image_build.yml

@@ -19,11 +19,11 @@ on:
         type: string
       cuda-major-version:
         description: 'CUDA major version'
-        default: "11"
+        default: "12"
         type: string
       cuda-minor-version:
         description: 'CUDA minor version'
-        default: "7"
+        default: "4"
         type: string
       platforms:
         description: 'Platforms'
@@ -324,7 +324,7 @@ jobs:
           docker pull quay.io/go-skynet/local-ai:${{ steps.meta_aio.outputs.version }}
           docker tag quay.io/go-skynet/local-ai:${{ steps.meta_aio.outputs.version }} quay.io/go-skynet/local-ai:${{ inputs.latest-image-aio }}
           docker push quay.io/go-skynet/local-ai:${{ inputs.latest-image-aio }}
-  
+
       - name: job summary
         run: |
           echo "Built image: ${{ steps.meta.outputs.labels }}" >> $GITHUB_STEP_SUMMARY

.github/workflows/notify-models.yaml

@@ -0,0 +1,170 @@
+name: Notifications for new models
+on:
+  pull_request:
+     types:
+       - closed
+
+jobs:
+  notify-discord:
+    if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
+    env:
+        MODEL_NAME: hermes-2-theta-llama-3-8b
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        fetch-depth: 0 # needed to checkout all branches for this Action to work
+    - name: Start LocalAI
+      run: |
+        echo "Starting LocalAI..."
+        docker run -e -ti -d --name local-ai -p 8080:8080 localai/localai:master-ffmpeg-core run --debug $MODEL_NAME
+        until [ "`docker inspect -f {{.State.Health.Status}} local-ai`" == "healthy" ]; do echo "Waiting for container to be ready";  docker logs --tail 10 local-ai; sleep 2; done
+      # Check the PR diff using the current branch and the base branch of the PR
+    - uses: GrantBirki/git-diff-action@v2.7.0
+      id: git-diff-action
+      with:
+            json_diff_file_output: diff.json
+            raw_diff_file_output: diff.txt
+            file_output_only: "true"
+    - name: Summarize
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      id: summarize
+      run: |
+            input="$(cat $DIFF)"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "You are LocalAI-bot. Write a discord message to notify everyone about the new model from the git diff. Make it informal. An example can include: the URL of the model, the name, and a brief description of the model if exists. Also add an hint on how to install it in LocalAI and that can be browsed over https://models.localai.io. For example: local-ai run model_name_here"
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary="$(echo $response | jq -r '.choices[0].message.content')"
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            echo "Summary:"
+            echo "$summary"
+            echo "payload sent"
+            echo "$json_payload"
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+            docker logs --tail 10 local-ai
+    - name: Discord notification
+      env:
+        DISCORD_WEBHOOK: ${{ secrets.DISCORD_WEBHOOK_URL }}
+        DISCORD_USERNAME: "LocalAI-Bot"
+        DISCORD_AVATAR: "https://avatars.githubusercontent.com/u/139863280?v=4"
+      uses: Ilshidur/action-discord@master
+      with:
+        args: ${{ steps.summarize.outputs.message }}
+    - name: Setup tmate session if fails
+      if: ${{ failure() }}
+      uses: mxschmitt/action-tmate@v3.18
+      with:
+        detached: true
+        connect-timeout-seconds: 180
+        limit-access-to-actor: true
+  notify-twitter:
+    if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
+    env:
+        MODEL_NAME: hermes-2-theta-llama-3-8b
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        fetch-depth: 0 # needed to checkout all branches for this Action to work
+    - name: Start LocalAI
+      run: |
+        echo "Starting LocalAI..."
+        docker run -e -ti -d --name local-ai -p 8080:8080 localai/localai:master-ffmpeg-core run --debug $MODEL_NAME
+        until [ "`docker inspect -f {{.State.Health.Status}} local-ai`" == "healthy" ]; do echo "Waiting for container to be ready";  docker logs --tail 10 local-ai; sleep 2; done
+      # Check the PR diff using the current branch and the base branch of the PR
+    - uses: GrantBirki/git-diff-action@v2.7.0
+      id: git-diff-action
+      with:
+            json_diff_file_output: diff.json
+            raw_diff_file_output: diff.txt
+            file_output_only: "true"
+    - name: Summarize
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      id: summarize
+      run: |
+            input="$(cat $DIFF)"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "You are LocalAI-bot. Write a twitter message to notify everyone about the new model from the git diff. Make it informal and really short. An example can include: the name, and a brief description of the model if exists. Also add an hint on how to install it in LocalAI. For example: local-ai run model_name_here"
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary="$(echo $response | jq -r '.choices[0].message.content')"
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            echo "Summary:"
+            echo "$summary"
+            echo "payload sent"
+            echo "$json_payload"
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+            docker logs --tail 10 local-ai
+    - uses: Eomm/why-don-t-you-tweet@v2
+      with:
+        tweet-message: ${{ steps.summarize.outputs.message }}
+      env:
+        # Get your tokens from https://developer.twitter.com/apps
+        TWITTER_CONSUMER_API_KEY: ${{ secrets.TWITTER_APP_KEY }}
+        TWITTER_CONSUMER_API_SECRET: ${{ secrets.TWITTER_APP_SECRET }}
+        TWITTER_ACCESS_TOKEN: ${{ secrets.TWITTER_ACCESS_TOKEN }}
+        TWITTER_ACCESS_TOKEN_SECRET: ${{ secrets.TWITTER_ACCESS_TOKEN_SECRET }}
+    - name: Setup tmate session if fails
+      if: ${{ failure() }}
+      uses: mxschmitt/action-tmate@v3.18
+      with:
+        detached: true
+        connect-timeout-seconds: 180
+        limit-access-to-actor: true

.github/workflows/notify-releases.yaml

@@ -0,0 +1,65 @@
+name: Release notifications
+on:
+  release:
+    types:
+      - published
+
+jobs:
+  notify-discord:
+    runs-on: ubuntu-latest
+    env:
+        RELEASE_BODY: ${{ github.event.release.body }}
+        RELEASE_TITLE: ${{ github.event.release.name }}
+        RELEASE_TAG_NAME: ${{ github.event.release.tag_name }}
+    steps:
+    - name: Start LocalAI
+      run: |
+        echo "Starting LocalAI..."
+        docker run -e -ti -d --name local-ai -p 8080:8080 localai/localai:master-ffmpeg-core run --debug $MODEL_NAME
+        until [ "`docker inspect -f {{.State.Health.Status}} local-ai`" == "healthy" ]; do echo "Waiting for container to be ready";  docker logs --tail 10 local-ai; sleep 2; done
+    - name: Summarize
+      id: summarize
+      run: |
+            input="$RELEASE_TITLE\b$RELEASE_BODY"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "Write a discord message with a bullet point summary of the release notes."
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI API
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary=$(echo $response | jq -r '.choices[0].message.content')
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+    - name: Discord notification
+      env:
+        DISCORD_WEBHOOK: ${{ secrets.DISCORD_WEBHOOK_URL_RELEASE }}
+        DISCORD_USERNAME: "LocalAI-Bot"
+        DISCORD_AVATAR: "https://avatars.githubusercontent.com/u/139863280?v=4"
+      uses: Ilshidur/action-discord@master
+      with:
+        args: ${{ steps.summarize.outputs.message }}

.github/workflows/release.yaml

@@ -31,7 +31,7 @@ jobs:
       - name: Dependencies
         run: |
           sudo apt-get update
-          sudo apt-get install build-essential ffmpeg protobuf-compiler ccache
+          sudo apt-get install build-essential ffmpeg protobuf-compiler ccache gawk
           sudo apt-get install -qy binutils-aarch64-linux-gnu gcc-aarch64-linux-gnu g++-aarch64-linux-gnu
       - name: Install CUDA Dependencies
         run: |
@@ -77,6 +77,16 @@ jobs:
             echo "set(CMAKE_FIND_ROOT_PATH_MODE_INCLUDE ONLY)" >> $CMAKE_CROSS_TOOLCHAIN && \
             echo "set(CMAKE_FIND_ROOT_PATH_MODE_PACKAGE ONLY)" >> $CMAKE_CROSS_TOOLCHAIN
           GRPC_DIR=$PWD/grpc
+
+          # http://google.github.io/googletest/quickstart-cmake.html
+          # Seems otherwise cross-arch fails to find it
+          echo "include(FetchContent)" >> $GRPC_DIR/CMakeLists.txt
+          echo "FetchContent_Declare(" >> $GRPC_DIR/CMakeLists.txt
+          echo "  googletest" >> $GRPC_DIR/CMakeLists.txt
+          echo "  URL https://github.com/google/googletest/archive/03597a01ee50ed33e9dfd640b249b4be3799d395.zip" >> $GRPC_DIR/CMakeLists.txt
+          echo ")" >> $GRPC_DIR/CMakeLists.txt
+          echo "FetchContent_MakeAvailable(googletest)" >> $GRPC_DIR/CMakeLists.txt
+
           cd grpc && cd cmake/build && sudo make --jobs 5 --output-sync=target install && \
           GRPC_CROSS_BUILD_DIR=$GRPC_DIR/cmake/cross_build && \
           mkdir -p $GRPC_CROSS_BUILD_DIR && \
@@ -96,11 +106,18 @@ jobs:
           CROSS_TOOLCHAIN=/usr/$GNU_HOST
           CROSS_STAGING_PREFIX=$CROSS_TOOLCHAIN/stage
           CMAKE_CROSS_TOOLCHAIN=/tmp/arm.toolchain.cmake
+
           go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
           go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
           export PATH=$PATH:$GOPATH/bin
           export PATH=/usr/local/cuda/bin:$PATH
-          GO_TAGS=p2p GOOS=linux GOARCH=arm64 CMAKE_ARGS="-DProtobuf_INCLUDE_DIRS=$CROSS_STAGING_PREFIX/include -DProtobuf_DIR=$CROSS_STAGING_PREFIX/lib/cmake/protobuf -DgRPC_DIR=$CROSS_STAGING_PREFIX/lib/cmake/grpc -DCMAKE_TOOLCHAIN_FILE=$CMAKE_CROSS_TOOLCHAIN -DCMAKE_C_COMPILER=aarch64-linux-gnu-gcc -DCMAKE_CXX_COMPILER=aarch64-linux-gnu-g++" make dist-cross-linux-arm64
+          sudo rm -rf /usr/aarch64-linux-gnu/lib/libstdc++.so.6
+          sudo cp -rf /usr/aarch64-linux-gnu/lib/libstdc++.so* /usr/aarch64-linux-gnu/lib/libstdc++.so.6
+          sudo cp /usr/aarch64-linux-gnu/lib/ld-linux-aarch64.so.1 ld.so
+          BACKEND_LIBS="./grpc/cmake/cross_build/third_party/re2/libre2.a ./grpc/cmake/cross_build/libgrpc.a ./grpc/cmake/cross_build/libgrpc++.a ./grpc/cmake/cross_build/third_party/protobuf/libprotobuf.a /usr/aarch64-linux-gnu/lib/libc.so.6 /usr/aarch64-linux-gnu/lib/libstdc++.so.6 /usr/aarch64-linux-gnu/lib/libgomp.so.1 /usr/aarch64-linux-gnu/lib/libm.so.6 /usr/aarch64-linux-gnu/lib/libgcc_s.so.1 /usr/aarch64-linux-gnu/lib/libdl.so.2 /usr/aarch64-linux-gnu/lib/libpthread.so.0 ./ld.so" \
+          GOOS=linux \
+          GOARCH=arm64 \
+          CMAKE_ARGS="-DProtobuf_INCLUDE_DIRS=$CROSS_STAGING_PREFIX/include -DProtobuf_DIR=$CROSS_STAGING_PREFIX/lib/cmake/protobuf -DgRPC_DIR=$CROSS_STAGING_PREFIX/lib/cmake/grpc -DCMAKE_TOOLCHAIN_FILE=$CMAKE_CROSS_TOOLCHAIN -DCMAKE_C_COMPILER=aarch64-linux-gnu-gcc -DCMAKE_CXX_COMPILER=aarch64-linux-gnu-g++" make dist-cross-linux-arm64
       - uses: actions/upload-artifact@v4
         with:
           name: LocalAI-linux-arm64
@@ -111,7 +128,13 @@ jobs:
         with:
           files: |
             release/*
-
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.18
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
   build-linux:
     runs-on: arc-runner-set
     steps:
@@ -134,7 +157,7 @@ jobs:
       - name: Dependencies
         run: |
           sudo apt-get update
-          sudo apt-get install -y wget curl build-essential ffmpeg protobuf-compiler ccache cmake
+          sudo apt-get install -y wget curl build-essential ffmpeg protobuf-compiler ccache gawk cmake
       - name: Intel Dependencies
         run: |
           wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | sudo tee /usr/share/keyrings/oneapi-archive-keyring.gpg > /dev/null
@@ -154,15 +177,15 @@ jobs:
           ROCM_VERSION: "6.1"
           AMDGPU_VERSION: "6.1"
         run: |
-            set -ex 
+            set -ex
 
             sudo apt-get update
-            sudo DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends ca-certificates curl libnuma-dev gnupg 
-            
-            curl -sL https://repo.radeon.com/rocm/rocm.gpg.key | sudo apt-key add - 
-              
+            sudo DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends ca-certificates curl libnuma-dev gnupg
+
+            curl -sL https://repo.radeon.com/rocm/rocm.gpg.key | sudo apt-key add -
+
             printf "deb [arch=amd64] https://repo.radeon.com/rocm/apt/$ROCM_VERSION/ jammy main" | sudo tee /etc/apt/sources.list.d/rocm.list
-            
+
             printf "deb [arch=amd64] https://repo.radeon.com/amdgpu/$AMDGPU_VERSION/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/amdgpu.list
             printf 'Package: *\nPin: release o=repo.radeon.com\nPin-Priority: 600' | sudo tee /etc/apt/preferences.d/rocm-pin-600
             sudo apt-get update
@@ -170,10 +193,10 @@ jobs:
             sudo DEBIAN_FRONTEND=noninteractive apt-get install -y \
                 hipblas-dev rocm-dev \
                 rocblas-dev
-          
+
             sudo apt-get clean
             sudo rm -rf /var/lib/apt/lists/*
-            sudo ldconfig 
+            sudo ldconfig
       - name: Cache grpc
         id: cache-grpc
         uses: actions/cache@v4
@@ -190,6 +213,7 @@ jobs:
       - name: Install gRPC
         run: |
           cd grpc && cd cmake/build && sudo make --jobs 5 --output-sync=target install
+      # BACKEND_LIBS needed for gpu-workload: /opt/intel/oneapi/*/lib/libiomp5.so /opt/intel/oneapi/*/lib/libmkl_core.so /opt/intel/oneapi/*/lib/libmkl_core.so.2 /opt/intel/oneapi/*/lib/libmkl_intel_ilp64.so /opt/intel/oneapi/*/lib/libmkl_intel_ilp64.so.2 /opt/intel/oneapi/*/lib/libmkl_sycl_blas.so /opt/intel/oneapi/*/lib/libmkl_sycl_blas.so.4 /opt/intel/oneapi/*/lib/libmkl_tbb_thread.so /opt/intel/oneapi/*/lib/libmkl_tbb_thread.so.2 /opt/intel/oneapi/*/lib/libsycl.so /opt/intel/oneapi/*/lib/libsycl.so.7 /opt/intel/oneapi/*/lib/libsycl.so.7.1.0 /opt/rocm-*/lib/libamdhip64.so /opt/rocm-*/lib/libamdhip64.so.5 /opt/rocm-*/lib/libamdhip64.so.6 /opt/rocm-*/lib/libamdhip64.so.6.1.60100 /opt/rocm-*/lib/libhipblas.so /opt/rocm-*/lib/libhipblas.so.2 /opt/rocm-*/lib/libhipblas.so.2.1.60100 /opt/rocm-*/lib/librocblas.so /opt/rocm-*/lib/librocblas.so.4 /opt/rocm-*/lib/librocblas.so.4.1.60100 /usr/lib/x86_64-linux-gnu/libstdc++.so.6 /usr/lib/x86_64-linux-gnu/libOpenCL.so.1 /usr/lib/x86_64-linux-gnu/libOpenCL.so.1.0.0 /usr/lib/x86_64-linux-gnu/libm.so.6 /usr/lib/x86_64-linux-gnu/libgcc_s.so.1 /usr/lib/x86_64-linux-gnu/libc.so.6 /usr/lib/x86_64-linux-gnu/librt.so.1 /usr/local/cuda-*/targets/x86_64-linux/lib/libcublas.so /usr/local/cuda-*/targets/x86_64-linux/lib/libcublasLt.so /usr/local/cuda-*/targets/x86_64-linux/lib/libcudart.so /usr/local/cuda-*/targets/x86_64-linux/lib/stubs/libcuda.so
       - name: Build
         id: build
         run: |
@@ -199,7 +223,9 @@ jobs:
           export PATH=/usr/local/cuda/bin:$PATH
           export PATH=/opt/rocm/bin:$PATH
           source /opt/intel/oneapi/setvars.sh
-          GO_TAGS=p2p make -j4 dist
+          sudo cp /lib64/ld-linux-x86-64.so.2 ld.so
+          BACKEND_LIBS="./ld.so ./sources/go-piper/piper/build/fi/lib/libfmt.a ./sources/go-piper/piper-phonemize/pi/lib/libonnxruntime.so.1.14.1 ./sources/go-piper/piper-phonemize/pi/src/libespeak-ng/libespeak-ng.so /usr/lib/x86_64-linux-gnu/libdl.so.2 /usr/lib/x86_64-linux-gnu/librt.so.1 /usr/lib/x86_64-linux-gnu/libpthread.so.0 ./sources/go-piper/piper-phonemize/pi/lib/libpiper_phonemize.so.1 ./sources/go-piper/piper/build/si/lib/libspdlog.a ./sources/go-piper/espeak/ei/lib/libucd.so" \
+          make -j4 dist
       - uses: actions/upload-artifact@v4
         with:
           name: LocalAI-linux
@@ -210,7 +236,13 @@ jobs:
         with:
           files: |
             release/*
-
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.18
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
   build-stablediffusion:
     runs-on: ubuntu-latest
     steps:
@@ -246,6 +278,48 @@ jobs:
           files: |
             release/*
 
+  build-macOS-x86_64:
+    runs-on: macos-13
+    steps:
+      - name: Clone
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - uses: actions/setup-go@v5
+        with:
+          go-version: '1.21.x'
+          cache: false
+      - name: Dependencies
+        run: |
+          brew install protobuf grpc
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
+      - name: Build
+        id: build
+        run: |
+          export C_INCLUDE_PATH=/usr/local/include
+          export CPLUS_INCLUDE_PATH=/usr/local/include
+          export PATH=$PATH:$GOPATH/bin
+
+          make dist
+      - uses: actions/upload-artifact@v4
+        with:
+          name: LocalAI-MacOS-x86_64
+          path: release/
+      - name: Release
+        uses: softprops/action-gh-release@v2
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: |
+            release/*
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.18
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true
+
   build-macOS-arm64:
     runs-on: macos-14
     steps:
@@ -268,7 +342,8 @@ jobs:
           export C_INCLUDE_PATH=/usr/local/include
           export CPLUS_INCLUDE_PATH=/usr/local/include
           export PATH=$PATH:$GOPATH/bin
-          GO_TAGS=p2p make dist
+
+          make dist
       - uses: actions/upload-artifact@v4
         with:
           name: LocalAI-MacOS-arm64
@@ -279,3 +354,10 @@ jobs:
         with:
           files: |
             release/*
+      - name: Setup tmate session if tests fail
+        if: ${{ failure() }}
+        uses: mxschmitt/action-tmate@v3.18
+        with:
+          detached: true
+          connect-timeout-seconds: 180
+          limit-access-to-actor: true

.github/workflows/test.yml

@@ -220,7 +220,7 @@ jobs:
           export CPLUS_INCLUDE_PATH=/usr/local/include
           # Used to run the newer GNUMake version from brew that supports --output-sync
           export PATH="/opt/homebrew/opt/make/libexec/gnubin:$PATH"
-          BUILD_TYPE="GITHUB_CI_HAS_BROKEN_METAL" CMAKE_ARGS="-DLLAMA_F16C=OFF -DLLAMA_AVX512=OFF -DLLAMA_AVX2=OFF -DLLAMA_FMA=OFF" make --jobs 4 --output-sync=target test
+          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@v3.18

Dockerfile

@@ -33,7 +33,7 @@ RUN curl -L -s https://go.dev/dl/go${GO_VERSION}.linux-${TARGETARCH}.tar.gz | ta
 ENV PATH $PATH:/root/go/bin:/usr/local/go/bin
 
 # Install grpc compilers
-RUN go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.1 && \
+RUN go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2 && \
     go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
 
 COPY --chmod=644 custom-ca-certs/* /usr/local/share/ca-certificates/
@@ -98,11 +98,27 @@ RUN pip install --user grpcio-tools
 FROM requirements-${IMAGE_TYPE} AS requirements-drivers
 
 ARG BUILD_TYPE
-ARG CUDA_MAJOR_VERSION=11
-ARG CUDA_MINOR_VERSION=8
+ARG CUDA_MAJOR_VERSION=12
+ARG CUDA_MINOR_VERSION=4
 
 ENV BUILD_TYPE=${BUILD_TYPE}
 
+# Vulkan requirements
+RUN <<EOT bash
+    if [ "${BUILD_TYPE}" = "vulkan" ]; then
+        apt-get update && \
+        apt-get install -y  --no-install-recommends \
+                        software-properties-common pciutils wget gpg-agent && \
+        wget -qO - https://packages.lunarg.com/lunarg-signing-key-pub.asc | apt-key add - && \
+        wget -qO /etc/apt/sources.list.d/lunarg-vulkan-jammy.list https://packages.lunarg.com/vulkan/lunarg-vulkan-jammy.list && \
+        apt-get update && \
+            apt-get install -y \
+            vulkan-sdk && \
+        apt-get clean && \
+        rm -rf /var/lib/apt/lists/*
+    fi
+EOT
+
 # CuBLAS requirements
 RUN <<EOT bash
     if [ "${BUILD_TYPE}" = "cublas" ]; then
@@ -266,6 +282,8 @@ COPY --from=grpc /opt/grpc /usr/local
 
 # Rebuild with defaults backends
 WORKDIR /build
+
+## Build the binary
 RUN make build
 
 RUN if [ ! -d "/build/sources/go-piper/piper-phonemize/pi/lib/" ]; then \
@@ -292,7 +310,7 @@ ENV REBUILD=false
 ENV HEALTHCHECK_ENDPOINT=http://localhost:8080/readyz
 ENV MAKEFLAGS=${MAKEFLAGS}
 
-ARG CUDA_MAJOR_VERSION=11
+ARG CUDA_MAJOR_VERSION=12
 ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
 ENV NVIDIA_REQUIRE_CUDA="cuda>=${CUDA_MAJOR_VERSION}.0"
 ENV NVIDIA_VISIBLE_DEVICES=all

Makefile

@@ -3,9 +3,11 @@ GOTEST=$(GOCMD) test
 GOVET=$(GOCMD) vet
 BINARY_NAME=local-ai
 
+DETECT_LIBS?=true
+
 # llama.cpp versions
 GOLLAMA_STABLE_VERSION?=2b57a8ae43e4699d3dc5d1496a1ccd42922993be
-CPPLLAMA_VERSION?=172c8256840ffd882ab9992ecedbb587d9b21f15
+CPPLLAMA_VERSION?=c4dd11d1d3903e1922c06242e189f6310fc4d8c3
 
 # gpt4all version
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
@@ -16,7 +18,7 @@ RWKV_REPO?=https://github.com/donomii/go-rwkv.cpp
 RWKV_VERSION?=661e7ae26d442f5cfebd2a0881b44e8c55949ec6
 
 # whisper.cpp version
-WHISPER_CPP_VERSION?=b29b3b29240aac8b71ce8e5a4360c1f1562ad66f
+WHISPER_CPP_VERSION?=1c31f9d4a8936aec550e6c4dc9ca5cae3b4f304a
 
 # bert.cpp version
 BERT_VERSION?=710044b124545415f555e4260d16b146c725a6e4
@@ -33,9 +35,11 @@ TINYDREAM_VERSION?=c04fa463ace9d9a6464313aa5f9cd0f953b6c057
 export BUILD_TYPE?=
 export STABLE_BUILD_TYPE?=$(BUILD_TYPE)
 export CMAKE_ARGS?=
+export BACKEND_LIBS?=
 
 CGO_LDFLAGS?=
 CGO_LDFLAGS_WHISPER?=
+CGO_LDFLAGS_WHISPER+=-lggml
 CUDA_LIBPATH?=/usr/local/cuda/lib64/
 GO_TAGS?=
 BUILD_ID?=
@@ -49,12 +53,12 @@ RANDOM := $(shell bash -c 'echo $$RANDOM')
 VERSION?=$(shell git describe --always --tags || echo "dev" )
 # go tool nm ./local-ai | grep Commit
 LD_FLAGS?=
-override LD_FLAGS += -X "github.com/go-skynet/LocalAI/internal.Version=$(VERSION)"
-override LD_FLAGS += -X "github.com/go-skynet/LocalAI/internal.Commit=$(shell git rev-parse HEAD)"
+override LD_FLAGS += -X "github.com/mudler/LocalAI/internal.Version=$(VERSION)"
+override LD_FLAGS += -X "github.com/mudler/LocalAI/internal.Commit=$(shell git rev-parse HEAD)"
 
 OPTIONAL_TARGETS?=
 
-OS := $(shell uname -s)
+export OS := $(shell uname -s)
 ARCH := $(shell uname -m)
 GREEN  := $(shell tput -Txterm setaf 2)
 YELLOW := $(shell tput -Txterm setaf 3)
@@ -80,29 +84,42 @@ ifeq ($(OS),Darwin)
 		BUILD_TYPE=metal
 	# disable metal if on Darwin and any other value is explicitly passed.
 	else ifneq ($(BUILD_TYPE),metal)
-		CMAKE_ARGS+=-DLLAMA_METAL=OFF
-		export LLAMA_NO_ACCELERATE=1
+		CMAKE_ARGS+=-DGGML_METAL=OFF
+		export GGML_NO_ACCELERATE=1
+		export GGML_NO_METAL=1
 	endif
 
 	ifeq ($(BUILD_TYPE),metal)
 #			-lcblas 	removed: it seems to always be listed as a duplicate flag.
 		CGO_LDFLAGS += -framework Accelerate
 	endif
+else
+CGO_LDFLAGS_WHISPER+=-lgomp
 endif
 
 ifeq ($(BUILD_TYPE),openblas)
 	CGO_LDFLAGS+=-lopenblas
-	export WHISPER_OPENBLAS=1
+	export GGML_OPENBLAS=1
 endif
 
-
 ifeq ($(BUILD_TYPE),cublas)
 	CGO_LDFLAGS+=-lcublas -lcudart -L$(CUDA_LIBPATH)
-	export LLAMA_CUBLAS=1
-	export WHISPER_CUDA=1
+	export GGML_CUDA=1
 	CGO_LDFLAGS_WHISPER+=-L$(CUDA_LIBPATH)/stubs/ -lcuda -lcufft
 endif
 
+ifeq ($(BUILD_TYPE),vulkan)
+	CMAKE_ARGS+=-DGGML_VULKAN=1
+endif
+
+ifneq (,$(findstring sycl,$(BUILD_TYPE)))
+	export GGML_SYCL=1
+endif
+
+ifeq ($(BUILD_TYPE),sycl_f16)
+	export GGML_SYCL_F16=1
+endif
+
 ifeq ($(BUILD_TYPE),hipblas)
 	ROCM_HOME ?= /opt/rocm
 	ROCM_PATH ?= /opt/rocm
@@ -111,27 +128,26 @@ ifeq ($(BUILD_TYPE),hipblas)
 	export CC=$(ROCM_HOME)/llvm/bin/clang
 	# llama-ggml has no hipblas support, so override it here.
 	export STABLE_BUILD_TYPE=
-	export WHISPER_HIPBLAS=1
+	export GGML_HIPBLAS=1
 	GPU_TARGETS ?= gfx900,gfx906,gfx908,gfx940,gfx941,gfx942,gfx90a,gfx1030,gfx1031,gfx1100,gfx1101
 	AMDGPU_TARGETS ?= "$(GPU_TARGETS)"
-	CMAKE_ARGS+=-DLLAMA_HIPBLAS=ON -DAMDGPU_TARGETS="$(AMDGPU_TARGETS)" -DGPU_TARGETS="$(GPU_TARGETS)"
+	CMAKE_ARGS+=-DGGML_HIPBLAS=ON -DAMDGPU_TARGETS="$(AMDGPU_TARGETS)" -DGPU_TARGETS="$(GPU_TARGETS)"
 	CGO_LDFLAGS += -O3 --rtlib=compiler-rt -unwindlib=libgcc -lhipblas -lrocblas --hip-link -L${ROCM_HOME}/lib/llvm/lib
 endif
 
 ifeq ($(BUILD_TYPE),metal)
 	CGO_LDFLAGS+=-framework Foundation -framework Metal -framework MetalKit -framework MetalPerformanceShaders
-	export LLAMA_METAL=1
-	export WHISPER_METAL=1
+	export GGML_METAL=1
 endif
 
 ifeq ($(BUILD_TYPE),clblas)
 	CGO_LDFLAGS+=-lOpenCL -lclblast
-	export WHISPER_CLBLAST=1
+	export GGML_OPENBLAS=1
 endif
 
 # glibc-static or glibc-devel-static required
 ifeq ($(STATIC),true)
-	LD_FLAGS=-linkmode external -extldflags -static
+	LD_FLAGS+=-linkmode external -extldflags -static
 endif
 
 ifeq ($(findstring stablediffusion,$(GO_TAGS)),stablediffusion)
@@ -165,6 +181,8 @@ ALL_GRPC_BACKENDS+=backend-assets/grpc/rwkv
 ALL_GRPC_BACKENDS+=backend-assets/grpc/whisper
 ALL_GRPC_BACKENDS+=backend-assets/grpc/local-store
 ALL_GRPC_BACKENDS+=$(OPTIONAL_GRPC)
+# Use filter-out to remove the specified backends
+ALL_GRPC_BACKENDS := $(filter-out $(SKIP_GRPC_BACKEND),$(ALL_GRPC_BACKENDS))
 
 GRPC_BACKENDS?=$(ALL_GRPC_BACKENDS) $(OPTIONAL_GRPC)
 TEST_PATHS?=./api/... ./pkg/... ./core/...
@@ -244,7 +262,7 @@ sources/whisper.cpp:
 	cd sources/whisper.cpp && git checkout -b build $(WHISPER_CPP_VERSION) && git submodule update --init --recursive --depth 1
 
 sources/whisper.cpp/libwhisper.a: sources/whisper.cpp
-	cd sources/whisper.cpp && $(MAKE) libwhisper.a
+	cd sources/whisper.cpp && $(MAKE) libwhisper.a libggml.a
 
 get-sources: sources/go-llama.cpp sources/gpt4all sources/go-piper sources/go-rwkv.cpp sources/whisper.cpp sources/go-bert.cpp sources/go-stable-diffusion sources/go-tiny-dream
 
@@ -313,6 +331,10 @@ build: prepare backend-assets grpcs ## Build the project
 	$(info ${GREEN}I BUILD_TYPE: ${YELLOW}$(BUILD_TYPE)${RESET})
 	$(info ${GREEN}I GO_TAGS: ${YELLOW}$(GO_TAGS)${RESET})
 	$(info ${GREEN}I LD_FLAGS: ${YELLOW}$(LD_FLAGS)${RESET})
+ifneq ($(BACKEND_LIBS),)
+	$(MAKE) backend-assets/lib
+	cp -f $(BACKEND_LIBS) backend-assets/lib/
+endif
 	CGO_LDFLAGS="$(CGO_LDFLAGS)" $(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o $(BINARY_NAME) ./
 
 build-minimal:
@@ -321,8 +343,14 @@ build-minimal:
 build-api:
 	BUILD_GRPC_FOR_BACKEND_LLAMA=true BUILD_API_ONLY=true GO_TAGS=none $(MAKE) build
 
+backend-assets/lib:
+	mkdir -p backend-assets/lib
+
 dist:
-	STATIC=true $(MAKE) backend-assets/grpc/llama-cpp-avx2
+	$(MAKE) backend-assets/grpc/llama-cpp-avx2
+ifeq ($(DETECT_LIBS),true)
+	scripts/prepare-libs.sh backend-assets/grpc/llama-cpp-avx2
+endif
 ifeq ($(OS),Darwin)
 	$(info ${GREEN}I Skip CUDA/hipblas build on MacOS${RESET})
 else
@@ -331,7 +359,11 @@ else
 	$(MAKE) backend-assets/grpc/llama-cpp-sycl_f16
 	$(MAKE) backend-assets/grpc/llama-cpp-sycl_f32
 endif
-	$(MAKE) build
+	GO_TAGS="tts p2p" $(MAKE) build
+ifeq ($(DETECT_LIBS),true)
+	scripts/prepare-libs.sh backend-assets/grpc/piper
+endif
+	GO_TAGS="tts p2p" STATIC=true $(MAKE) build
 	mkdir -p release
 # if BUILD_ID is empty, then we don't append it to the binary name
 ifeq ($(BUILD_ID),)
@@ -342,9 +374,9 @@ else
 	shasum -a 256 release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-$(ARCH) > release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-$(ARCH).sha256
 endif
 
-dist-cross-linux-arm64: 
-	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_NATIVE=off" GRPC_BACKENDS="backend-assets/grpc/llama-cpp-fallback backend-assets/grpc/llama-cpp-grpc backend-assets/util/llama-cpp-rpc-server" \
-	$(MAKE) build
+dist-cross-linux-arm64:
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_NATIVE=off" GRPC_BACKENDS="backend-assets/grpc/llama-cpp-fallback backend-assets/grpc/llama-cpp-grpc backend-assets/util/llama-cpp-rpc-server" \
+	STATIC=true $(MAKE) build
 	mkdir -p release
 # if BUILD_ID is empty, then we don't append it to the binary name
 ifeq ($(BUILD_ID),)
@@ -393,7 +425,7 @@ prepare-e2e:
 	mkdir -p $(TEST_DIR)
 	cp -rfv $(abspath ./tests/e2e-fixtures)/gpu.yaml $(TEST_DIR)/gpu.yaml
 	test -e $(TEST_DIR)/ggllm-test-model.bin || wget -q https://huggingface.co/TheBloke/CodeLlama-7B-Instruct-GGUF/resolve/main/codellama-7b-instruct.Q2_K.gguf -O $(TEST_DIR)/ggllm-test-model.bin
-	docker build --build-arg GRPC_BACKENDS="$(GRPC_BACKENDS)" --build-arg IMAGE_TYPE=core --build-arg BUILD_TYPE=$(BUILD_TYPE) --build-arg CUDA_MAJOR_VERSION=11 --build-arg CUDA_MINOR_VERSION=7 --build-arg FFMPEG=true -t localai-tests .
+	docker build --build-arg GRPC_BACKENDS="$(GRPC_BACKENDS)" --build-arg IMAGE_TYPE=core --build-arg BUILD_TYPE=$(BUILD_TYPE) --build-arg CUDA_MAJOR_VERSION=12 --build-arg CUDA_MINOR_VERSION=4 --build-arg FFMPEG=true -t localai-tests .
 
 run-e2e-image:
 	ls -liah $(abspath ./tests/e2e-fixtures)
@@ -700,21 +732,21 @@ backend-assets/grpc/llama-cpp-avx2: backend-assets/grpc
 	cp -rf backend/cpp/llama backend/cpp/llama-avx2
 	$(MAKE) -C backend/cpp/llama-avx2 purge
 	$(info ${GREEN}I llama-cpp build info:avx2${RESET})
-	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=on -DLLAMA_AVX2=on -DLLAMA_AVX512=off -DLLAMA_FMA=on -DLLAMA_F16C=on" $(MAKE) VARIANT="llama-avx2" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=on -DGGML_AVX512=off -DGGML_FMA=on -DGGML_F16C=on" $(MAKE) VARIANT="llama-avx2" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-avx2/grpc-server backend-assets/grpc/llama-cpp-avx2
 
 backend-assets/grpc/llama-cpp-avx: backend-assets/grpc
 	cp -rf backend/cpp/llama backend/cpp/llama-avx
 	$(MAKE) -C backend/cpp/llama-avx purge
 	$(info ${GREEN}I llama-cpp build info:avx${RESET})
-	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=on -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off" $(MAKE) VARIANT="llama-avx" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" $(MAKE) VARIANT="llama-avx" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-avx/grpc-server backend-assets/grpc/llama-cpp-avx
 
 backend-assets/grpc/llama-cpp-fallback: backend-assets/grpc
 	cp -rf backend/cpp/llama backend/cpp/llama-fallback
 	$(MAKE) -C backend/cpp/llama-fallback purge
 	$(info ${GREEN}I llama-cpp build info:fallback${RESET})
-	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=off -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off" $(MAKE) VARIANT="llama-fallback" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=off -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" $(MAKE) VARIANT="llama-fallback" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-fallback/grpc-server backend-assets/grpc/llama-cpp-fallback
 # TODO: every binary should have its own folder instead, so can have different metal implementations
 ifeq ($(BUILD_TYPE),metal)
@@ -725,7 +757,7 @@ backend-assets/grpc/llama-cpp-cuda: backend-assets/grpc
 	cp -rf backend/cpp/llama backend/cpp/llama-cuda
 	$(MAKE) -C backend/cpp/llama-cuda purge
 	$(info ${GREEN}I llama-cpp build info:cuda${RESET})
-	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_AVX=on -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off -DLLAMA_CUDA=ON" $(MAKE) VARIANT="llama-cuda" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off -DGGML_CUDA=ON" $(MAKE) VARIANT="llama-cuda" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-cuda/grpc-server backend-assets/grpc/llama-cpp-cuda
 
 backend-assets/grpc/llama-cpp-hipblas: backend-assets/grpc
@@ -753,7 +785,7 @@ backend-assets/grpc/llama-cpp-grpc: backend-assets/grpc
 	cp -rf backend/cpp/llama backend/cpp/llama-grpc
 	$(MAKE) -C backend/cpp/llama-grpc purge
 	$(info ${GREEN}I llama-cpp build info:grpc${RESET})
-	CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_RPC=ON -DLLAMA_AVX=off -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off" $(MAKE) VARIANT="llama-grpc" build-llama-cpp-grpc-server
+	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_RPC=ON -DGGML_AVX=off -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" TARGET="--target grpc-server --target rpc-server" $(MAKE) VARIANT="llama-grpc" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-grpc/grpc-server backend-assets/grpc/llama-cpp-grpc
 
 backend-assets/util/llama-cpp-rpc-server: backend-assets/grpc/llama-cpp-grpc
@@ -781,7 +813,7 @@ backend-assets/grpc/tinydream: sources/go-tiny-dream sources/go-tiny-dream/libti
 	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/tinydream ./backend/go/image/tinydream
 
 backend-assets/grpc/whisper: sources/whisper.cpp sources/whisper.cpp/libwhisper.a backend-assets/grpc
-	CGO_LDFLAGS="$(CGO_LDFLAGS) $(CGO_LDFLAGS_WHISPER)" C_INCLUDE_PATH=$(CURDIR)/sources/whisper.cpp LIBRARY_PATH=$(CURDIR)/sources/whisper.cpp \
+	CGO_LDFLAGS="$(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/
 
 backend-assets/grpc/local-store: backend-assets/grpc
@@ -803,6 +835,17 @@ docker:
 		--build-arg BUILD_TYPE=$(BUILD_TYPE) \
 		-t $(DOCKER_IMAGE) .
 
+docker-cuda11:
+	docker build \
+		--build-arg CUDA_MAJOR_VERSION=11 \
+		--build-arg CUDA_MINOR_VERSION=8 \
+		--build-arg BASE_IMAGE=$(BASE_IMAGE) \
+		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
+		--build-arg GO_TAGS="$(GO_TAGS)" \
+		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
+		--build-arg BUILD_TYPE=$(BUILD_TYPE) \
+		-t $(DOCKER_IMAGE)-cuda11 .
+
 docker-aio:
 	@echo "Building AIO image with base $(BASE_IMAGE) as $(DOCKER_AIO_IMAGE)"
 	docker build \
@@ -839,7 +882,7 @@ gen-assets:
 	$(GOCMD) run core/dependencies_manager/manager.go embedded/webui_static.yaml core/http/static/assets
 
 ## Documentation
-docs/layouts/_default: 
+docs/layouts/_default:
 	mkdir -p docs/layouts/_default
 
 docs/static/gallery.html: docs/layouts/_default
@@ -854,4 +897,4 @@ docs-clean:
 
 .PHONY: docs
 docs: docs/static/gallery.html
-	cd docs && hugo serve
+	cd docs && hugo serve

README.md

@@ -48,6 +48,13 @@
 
 ![screen](https://github.com/mudler/LocalAI/assets/2420543/20b5ccd2-8393-44f0-aaf6-87a23806381e)
 
+Run the installer script:
+
+```bash
+curl https://localai.io/install.sh | sh
+```
+
+Or run with docker:
 ```bash
 docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-aio-cpu
 # Alternative images:
@@ -65,6 +72,8 @@ docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-aio-cpu
 
 [Roadmap](https://github.com/mudler/LocalAI/issues?q=is%3Aissue+is%3Aopen+label%3Aroadmap)
 
+- 🔥🔥 🆕 P2P Dashboard, LocalAI Federated mode and AI Swarms: https://github.com/mudler/LocalAI/pull/2723
+- 🆕 You can browse now the model gallery without LocalAI! Check out https://models.localai.io
 - 🔥🔥 Decentralized llama.cpp:  https://github.com/mudler/LocalAI/pull/2343 (peer2peer llama.cpp!) 👉 Docs  https://localai.io/features/distribute/
 - 🔥🔥 Openvoice: https://github.com/mudler/LocalAI/pull/2334
 - 🆕 Function calls without grammars and mixed mode: https://github.com/mudler/LocalAI/pull/2328
@@ -96,6 +105,7 @@ If you want to help and contribute, issues up for grabs: https://github.com/mudl
 - 🥽 [Vision API](https://localai.io/features/gpt-vision/)
 - 📈 [Reranker API](https://localai.io/features/reranker/)
 - 🆕🖧 [P2P Inferencing](https://localai.io/features/distribute/)
+- 🌍 Integrated WebUI!
 
 ## 💻 Usage
 

backend/backend.proto

@@ -230,6 +230,7 @@ message TranscriptRequest {
   string dst = 2;
   string language = 3;
   uint32 threads = 4;
+  bool translate = 5;
 }
 
 message TranscriptResult {

backend/cpp/llama/CMakeLists.txt

@@ -83,4 +83,11 @@ target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_
 target_compile_features(${TARGET} PRIVATE cxx_std_11)
 if(TARGET BUILD_INFO)
   add_dependencies(${TARGET} BUILD_INFO)
-endif()
+endif()
+
+include(FetchContent)
+FetchContent_Declare(
+  googletest
+  URL https://github.com/google/googletest/archive/03597a01ee50ed33e9dfd640b249b4be3799d395.zip
+)
+FetchContent_MakeAvailable(googletest)

backend/cpp/llama/Makefile

@@ -4,34 +4,44 @@ LLAMA_VERSION?=
 CMAKE_ARGS?=
 BUILD_TYPE?=
 ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh
+TARGET?=--target grpc-server
 
-# If build type is cublas, then we set -DLLAMA_CUBLAS=ON to CMAKE_ARGS automatically
+# 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 ($(BUILD_TYPE),cublas)
-	CMAKE_ARGS+=-DLLAMA_CUBLAS=ON
-# If build type is openblas then we set -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
+	CMAKE_ARGS+=-DGGML_CUDA=ON
+# If build type is openblas then we set -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
 # to CMAKE_ARGS automatically
 else ifeq ($(BUILD_TYPE),openblas)
-	CMAKE_ARGS+=-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
-# If build type is clblas (openCL) we set -DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
+	CMAKE_ARGS+=-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
+# If build type is clblas (openCL) we set -DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
 else ifeq ($(BUILD_TYPE),clblas)
-	CMAKE_ARGS+=-DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
+	CMAKE_ARGS+=-DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
 # If it's hipblas we do have also to set CC=/opt/rocm/llvm/bin/clang CXX=/opt/rocm/llvm/bin/clang++ 
 else ifeq ($(BUILD_TYPE),hipblas)
-	CMAKE_ARGS+=-DLLAMA_HIPBLAS=ON
-# If it's OSX, DO NOT embed the metal library - -DLLAMA_METAL_EMBED_LIBRARY=ON requires further investigation
+	CMAKE_ARGS+=-DGGML_HIPBLAS=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 ($(OS),darwin)
+else ifeq ($(OS),Darwin)
 	ifneq ($(BUILD_TYPE),metal)
-		CMAKE_ARGS+=-DLLAMA_METAL=OFF
+		CMAKE_ARGS+=-DGGML_METAL=OFF
+	else
+		CMAKE_ARGS+=-DGGML_METAL=ON
+# Until this is tested properly, we disable embedded metal file
+# as we already embed it as part of the LocalAI assets
+		CMAKE_ARGS+=-DGGML_METAL_EMBED_LIBRARY=OFF
+		TARGET+=--target ggml-metal
 	endif
 endif
 
 ifeq ($(BUILD_TYPE),sycl_f16)
-	CMAKE_ARGS+=-DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_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+=-DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
+	CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
 endif
 
 llama.cpp:
@@ -61,9 +71,9 @@ clean: purge
 grpc-server: llama.cpp llama.cpp/examples/grpc-server
 	@echo "Building grpc-server with $(BUILD_TYPE) build type and $(CMAKE_ARGS)"
 ifneq (,$(findstring sycl,$(BUILD_TYPE)))
-	bash -c "source $(ONEAPI_VARS); \
-	cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && $(MAKE)"
+	+bash -c "source $(ONEAPI_VARS); \
+	cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && cmake --build . --config Release $(TARGET)"
 else
-	cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && $(MAKE)
+	+cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && cmake --build . --config Release $(TARGET)
 endif
 	cp llama.cpp/build/bin/grpc-server .

backend/cpp/llama/grpc-server.cpp

@@ -886,6 +886,8 @@ struct llama_server_context
             {"task_id", slot->task_id},
         });
 
+        LOG_TEE("sampling: \n%s\n", llama_sampling_print(slot->sparams).c_str());
+
         return true;
     }
 

backend/go/image/stablediffusion/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/image/stablediffusion/stablediffusion.go

@@ -3,9 +3,9 @@ package main
 // This is a wrapper to statisfy the GRPC service interface
 // It is meant to be used by the main executable that is the server for the specific backend type (falcon, gpt3, etc)
 import (
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	"github.com/go-skynet/LocalAI/pkg/stablediffusion"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/stablediffusion"
 )
 
 type Image struct {

backend/go/image/tinydream/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/image/tinydream/tinydream.go

@@ -3,9 +3,9 @@ package main
 // This is a wrapper to statisfy the GRPC service interface
 // It is meant to be used by the main executable that is the server for the specific backend type (falcon, gpt3, etc)
 import (
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	"github.com/go-skynet/LocalAI/pkg/tinydream"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/tinydream"
 )
 
 type Image struct {

backend/go/llm/bert/bert.go

@@ -5,8 +5,8 @@ package main
 import (
 	bert "github.com/go-skynet/go-bert.cpp"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 )
 
 type Embeddings struct {

backend/go/llm/bert/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/llm/gpt4all/gpt4all.go

@@ -5,8 +5,8 @@ package main
 import (
 	"fmt"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 	gpt4all "github.com/nomic-ai/gpt4all/gpt4all-bindings/golang"
 )
 

backend/go/llm/gpt4all/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/llm/langchain/langchain.go

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

backend/go/llm/langchain/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/llm/llama-ggml/llama.go

@@ -5,9 +5,9 @@ package main
 import (
 	"fmt"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
 	"github.com/go-skynet/go-llama.cpp"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 )
 
 type LLM struct {

backend/go/llm/llama-ggml/main.go

@@ -3,7 +3,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/llm/llama/llama.go

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

backend/go/llm/llama/main.go

@@ -7,7 +7,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/llm/rwkv/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/llm/rwkv/rwkv.go

@@ -7,8 +7,8 @@ import (
 	"path/filepath"
 
 	"github.com/donomii/go-rwkv.cpp"
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 )
 
 const tokenizerSuffix = ".tokenizer.json"
@@ -31,7 +31,7 @@ func (llm *LLM) Load(opts *pb.ModelOptions) error {
 	model := rwkv.LoadFiles(opts.ModelFile, tokenizerPath, uint32(opts.GetThreads()))
 
 	if model == nil {
-		return fmt.Errorf("could not load model")
+		return fmt.Errorf("rwkv could not load model")
 	}
 	llm.rwkv = model
 	return nil

backend/go/stores/main.go

@@ -6,7 +6,7 @@ import (
 	"flag"
 	"os"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 	"github.com/rs/zerolog"
 	"github.com/rs/zerolog/log"
 )

backend/go/stores/store.go

@@ -8,8 +8,8 @@ import (
 	"math"
 	"slices"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 
 	"github.com/rs/zerolog/log"
 )

backend/go/transcribe/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/transcribe/transcript.go

@@ -8,7 +8,7 @@ import (
 
 	"github.com/ggerganov/whisper.cpp/bindings/go/pkg/whisper"
 	"github.com/go-audio/wav"
-	"github.com/go-skynet/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/schema"
 )
 
 func ffmpegCommand(args []string) (string, error) {
@@ -29,7 +29,7 @@ func audioToWav(src, dst string) error {
 	return nil
 }
 
-func Transcript(model whisper.Model, audiopath, language string, threads uint) (schema.TranscriptionResult, error) {
+func Transcript(model whisper.Model, audiopath, language string, translate bool, threads uint) (schema.TranscriptionResult, error) {
 	res := schema.TranscriptionResult{}
 
 	dir, err := os.MkdirTemp("", "whisper")
@@ -75,6 +75,10 @@ func Transcript(model whisper.Model, audiopath, language string, threads uint) (
 		context.SetLanguage("auto")
 	}
 
+	if translate {
+		context.SetTranslate(true)
+	}
+
 	if err := context.Process(data, nil, nil); err != nil {
 		return res, err
 	}

backend/go/transcribe/whisper.go

@@ -4,9 +4,9 @@ package main
 // It is meant to be used by the main executable that is the server for the specific backend type (falcon, gpt3, etc)
 import (
 	"github.com/ggerganov/whisper.cpp/bindings/go/pkg/whisper"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 )
 
 type Whisper struct {
@@ -22,5 +22,5 @@ func (sd *Whisper) Load(opts *pb.ModelOptions) error {
 }
 
 func (sd *Whisper) AudioTranscription(opts *pb.TranscriptRequest) (schema.TranscriptionResult, error) {
-	return Transcript(sd.whisper, opts.Dst, opts.Language, uint(opts.Threads))
+	return Transcript(sd.whisper, opts.Dst, opts.Language, opts.Translate, uint(opts.Threads))
 }

backend/go/tts/main.go

@@ -5,7 +5,7 @@ package main
 import (
 	"flag"
 
-	grpc "github.com/go-skynet/LocalAI/pkg/grpc"
+	grpc "github.com/mudler/LocalAI/pkg/grpc"
 )
 
 var (

backend/go/tts/piper.go

@@ -7,8 +7,8 @@ import (
 	"os"
 	"path/filepath"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/base"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 	piper "github.com/mudler/go-piper"
 )
 

backend/python/common/libbackend.sh

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

backend/python/diffusers/backend.py

@@ -17,7 +17,7 @@ import backend_pb2_grpc
 
 import grpc
 
-from diffusers import StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, EulerAncestralDiscreteScheduler
+from diffusers import StableDiffusion3Pipeline, StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, EulerAncestralDiscreteScheduler
 from diffusers import StableDiffusionImg2ImgPipeline, AutoPipelineForText2Image, ControlNetModel, StableVideoDiffusionPipeline
 from diffusers.pipelines.stable_diffusion import safety_checker
 from diffusers.utils import load_image,export_to_video
@@ -225,6 +225,17 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                         torch_dtype=torchType, 
                         use_safetensors=True, 
                         variant=variant)
+            elif request.PipelineType == "StableDiffusion3Pipeline":
+                if fromSingleFile:
+                    self.pipe = StableDiffusion3Pipeline.from_single_file(modelFile,
+                                                               torch_dtype=torchType,
+                                                               use_safetensors=True)
+                else:
+                    self.pipe = StableDiffusion3Pipeline.from_pretrained(
+                        request.Model, 
+                        torch_dtype=torchType, 
+                        use_safetensors=True, 
+                        variant=variant)
 
             if CLIPSKIP and request.CLIPSkip != 0:
                 self.clip_skip = request.CLIPSkip

backend/python/diffusers/requirements.txt

@@ -5,6 +5,7 @@ grpcio==1.64.0
 opencv-python
 pillow
 protobuf
+sentencepiece
 torch
 transformers
-certifi
+certifi

core/application.go

@@ -1,9 +1,9 @@
 package core
 
 import (
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/services"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/services"
+	"github.com/mudler/LocalAI/pkg/model"
 )
 
 // The purpose of this structure is to hold pointers to all initialized services, to make plumbing easy

core/backend/embeddings.go

@@ -3,10 +3,10 @@ package backend
 import (
 	"fmt"
 
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc"
-	model "github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/grpc"
+	model "github.com/mudler/LocalAI/pkg/model"
 )
 
 func ModelEmbedding(s string, tokens []int, loader *model.ModelLoader, backendConfig config.BackendConfig, appConfig *config.ApplicationConfig) (func() ([]float32, error), error) {

core/backend/image.go

@@ -1,10 +1,10 @@
 package backend
 
 import (
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	model "github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/grpc/proto"
+	model "github.com/mudler/LocalAI/pkg/model"
 )
 
 func ImageGeneration(height, width, mode, step, seed int, positive_prompt, negative_prompt, src, dst string, loader *model.ModelLoader, backendConfig config.BackendConfig, appConfig *config.ApplicationConfig) (func() error, error) {

core/backend/llm.go

@@ -9,14 +9,14 @@ import (
 	"sync"
 	"unicode/utf8"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
 
-	"github.com/go-skynet/LocalAI/pkg/gallery"
-	"github.com/go-skynet/LocalAI/pkg/grpc"
-	"github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	model "github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/go-skynet/LocalAI/pkg/utils"
+	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/pkg/grpc"
+	"github.com/mudler/LocalAI/pkg/grpc/proto"
+	model "github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/utils"
 )
 
 type LLMResponse struct {

core/backend/options.go

@@ -5,9 +5,9 @@ import (
 	"os"
 	"path/filepath"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/config"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
 
@@ -142,12 +142,14 @@ func gRPCPredictOpts(c config.BackendConfig, modelPath string) *pb.PredictOption
 		MirostatTAU:         float32(*c.LLMConfig.MirostatTAU),
 		Debug:               *c.Debug,
 		StopPrompts:         c.StopWords,
-		Repeat:              int32(c.RepeatPenalty),
+		Repeat:              int32(c.RepeatLastN),
+		FrequencyPenalty:    float32(c.FrequencyPenalty),
+		PresencePenalty:     float32(c.PresencePenalty),
+		Penalty:             float32(c.RepeatPenalty),
 		NKeep:               int32(c.Keep),
 		Batch:               int32(c.Batch),
 		IgnoreEOS:           c.IgnoreEOS,
 		Seed:                getSeed(c),
-		FrequencyPenalty:    float32(c.FrequencyPenalty),
 		MLock:               *c.MMlock,
 		MMap:                *c.MMap,
 		MainGPU:             c.MainGPU,

core/backend/rerank.go

@@ -4,9 +4,9 @@ import (
 	"context"
 	"fmt"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	model "github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/pkg/grpc/proto"
+	model "github.com/mudler/LocalAI/pkg/model"
 )
 
 func Rerank(backend, modelFile string, request *proto.RerankRequest, loader *model.ModelLoader, appConfig *config.ApplicationConfig, backendConfig config.BackendConfig) (*proto.RerankResult, error) {

core/backend/stores.go

@@ -1,10 +1,10 @@
 package backend
 
 import (
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/grpc"
+	"github.com/mudler/LocalAI/pkg/model"
 )
 
 func StoreBackend(sl *model.ModelLoader, appConfig *config.ApplicationConfig, storeName string) (grpc.Backend, error) {

core/backend/transcript.go

@@ -4,14 +4,14 @@ import (
 	"context"
 	"fmt"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	model "github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/grpc/proto"
+	model "github.com/mudler/LocalAI/pkg/model"
 )
 
-func ModelTranscription(audio, language string, ml *model.ModelLoader, backendConfig config.BackendConfig, appConfig *config.ApplicationConfig) (*schema.TranscriptionResult, error) {
+func ModelTranscription(audio, language string, translate bool, ml *model.ModelLoader, backendConfig config.BackendConfig, appConfig *config.ApplicationConfig) (*schema.TranscriptionResult, error) {
 
 	opts := modelOpts(backendConfig, appConfig, []model.Option{
 		model.WithBackendString(model.WhisperBackend),
@@ -31,8 +31,9 @@ func ModelTranscription(audio, language string, ml *model.ModelLoader, backendCo
 	}
 
 	return whisperModel.AudioTranscription(context.Background(), &proto.TranscriptRequest{
-		Dst:      audio,
-		Language: language,
-		Threads:  uint32(*backendConfig.Threads),
+		Dst:       audio,
+		Language:  language,
+		Translate: translate,
+		Threads:   uint32(*backendConfig.Threads),
 	})
 }

core/backend/tts.go

@@ -6,11 +6,11 @@ import (
 	"os"
 	"path/filepath"
 
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	model "github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/go-skynet/LocalAI/pkg/utils"
+	"github.com/mudler/LocalAI/pkg/grpc/proto"
+	model "github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/utils"
 )
 
 func generateUniqueFileName(dir, baseName, ext string) string {

core/cli/cli.go

@@ -1,14 +1,15 @@
 package cli
 
 import (
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
-	"github.com/go-skynet/LocalAI/core/cli/worker"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/core/cli/worker"
 )
 
 var CLI struct {
 	cliContext.Context `embed:""`
 
 	Run        RunCMD        `cmd:"" help:"Run LocalAI, this the default command if no other command is specified. Run 'local-ai run --help' for more information" default:"withargs"`
+	Federated  FederatedCLI  `cmd:"" help:"Run LocalAI in federated mode"`
 	Models     ModelsCMD     `cmd:"" help:"Manage LocalAI models and definitions"`
 	TTS        TTSCMD        `cmd:"" help:"Convert text to speech"`
 	Transcript TranscriptCMD `cmd:"" help:"Convert audio to text"`

core/cli/federated.go

@@ -0,0 +1,130 @@
+package cli
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"io"
+	"net"
+	"time"
+
+	"math/rand/v2"
+
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/core/p2p"
+	"github.com/mudler/edgevpn/pkg/node"
+	"github.com/mudler/edgevpn/pkg/protocol"
+	"github.com/mudler/edgevpn/pkg/types"
+	"github.com/rs/zerolog/log"
+)
+
+type FederatedCLI struct {
+	Address        string `env:"LOCALAI_ADDRESS,ADDRESS" default:":8080" help:"Bind address for the API server" group:"api"`
+	Peer2PeerToken string `env:"LOCALAI_P2P_TOKEN,P2P_TOKEN,TOKEN" name:"p2ptoken" help:"Token for P2P mode (optional)" group:"p2p"`
+}
+
+func (f *FederatedCLI) Run(ctx *cliContext.Context) error {
+
+	n, err := p2p.NewNode(f.Peer2PeerToken)
+	if err != nil {
+		return fmt.Errorf("creating a new node: %w", err)
+	}
+	err = n.Start(context.Background())
+	if err != nil {
+		return fmt.Errorf("creating a new node: %w", err)
+	}
+
+	if err := p2p.ServiceDiscoverer(context.Background(), n, f.Peer2PeerToken, p2p.FederatedID, nil); err != nil {
+		return err
+	}
+
+	return Proxy(context.Background(), n, f.Address, p2p.FederatedID)
+}
+
+func Proxy(ctx context.Context, node *node.Node, listenAddr, service string) error {
+
+	log.Info().Msgf("Allocating service '%s' on: %s", service, listenAddr)
+	// Open local port for listening
+	l, err := net.Listen("tcp", listenAddr)
+	if err != nil {
+		log.Error().Err(err).Msg("Error listening")
+		return err
+	}
+	//	ll.Info("Binding local port on", srcaddr)
+
+	ledger, _ := node.Ledger()
+
+	// Announce ourselves so nodes accepts our connection
+	ledger.Announce(
+		ctx,
+		10*time.Second,
+		func() {
+			// Retrieve current ID for ip in the blockchain
+			//_, found := ledger.GetKey(protocol.UsersLedgerKey, node.Host().ID().String())
+			// If mismatch, update the blockchain
+			//if !found {
+			updatedMap := map[string]interface{}{}
+			updatedMap[node.Host().ID().String()] = &types.User{
+				PeerID:    node.Host().ID().String(),
+				Timestamp: time.Now().String(),
+			}
+			ledger.Add(protocol.UsersLedgerKey, updatedMap)
+			//	}
+		},
+	)
+
+	defer l.Close()
+	for {
+		select {
+		case <-ctx.Done():
+			return errors.New("context canceled")
+		default:
+			log.Debug().Msg("New for connection")
+			// Listen for an incoming connection.
+			conn, err := l.Accept()
+			if err != nil {
+				fmt.Println("Error accepting: ", err.Error())
+				continue
+			}
+
+			// Handle connections in a new goroutine, forwarding to the p2p service
+			go func() {
+				var tunnelAddresses []string
+				for _, v := range p2p.GetAvailableNodes(p2p.FederatedID) {
+					if v.IsOnline() {
+						tunnelAddresses = append(tunnelAddresses, v.TunnelAddress)
+					} else {
+						log.Info().Msgf("Node %s is offline", v.ID)
+					}
+				}
+
+				// open a TCP stream to one of the tunnels
+				// chosen randomly
+				// TODO: optimize this and track usage
+				tunnelAddr := tunnelAddresses[rand.IntN(len(tunnelAddresses))]
+
+				tunnelConn, err := net.Dial("tcp", tunnelAddr)
+				if err != nil {
+					log.Error().Err(err).Msg("Error connecting to tunnel")
+					return
+				}
+
+				log.Info().Msgf("Redirecting %s to %s", conn.LocalAddr().String(), tunnelConn.RemoteAddr().String())
+				closer := make(chan struct{}, 2)
+				go copyStream(closer, tunnelConn, conn)
+				go copyStream(closer, conn, tunnelConn)
+				<-closer
+
+				tunnelConn.Close()
+				conn.Close()
+				//	ll.Infof("(service %s) Done handling %s", serviceID, l.Addr().String())
+			}()
+		}
+	}
+
+}
+
+func copyStream(closer chan struct{}, dst io.Writer, src io.Reader) {
+	defer func() { closer <- struct{}{} }() // connection is closed, send signal to stop proxy
+	io.Copy(dst, src)
+}

core/cli/models.go

@@ -4,10 +4,12 @@ import (
 	"encoding/json"
 	"fmt"
 
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/pkg/gallery"
-	"github.com/go-skynet/LocalAI/pkg/startup"
+	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/pkg/downloader"
+	"github.com/mudler/LocalAI/pkg/startup"
 	"github.com/rs/zerolog/log"
 	"github.com/schollz/progressbar/v3"
 )
@@ -33,7 +35,7 @@ type ModelsCMD struct {
 }
 
 func (ml *ModelsList) Run(ctx *cliContext.Context) error {
-	var galleries []gallery.Gallery
+	var galleries []config.Gallery
 	if err := json.Unmarshal([]byte(ml.Galleries), &galleries); err != nil {
 		log.Error().Err(err).Msg("unable to load galleries")
 	}
@@ -53,10 +55,11 @@ func (ml *ModelsList) Run(ctx *cliContext.Context) error {
 }
 
 func (mi *ModelsInstall) Run(ctx *cliContext.Context) error {
-	var galleries []gallery.Gallery
+	var galleries []config.Gallery
 	if err := json.Unmarshal([]byte(mi.Galleries), &galleries); err != nil {
 		log.Error().Err(err).Msg("unable to load galleries")
 	}
+
 	for _, modelName := range mi.ModelArgs {
 
 		progressBar := progressbar.NewOptions(
@@ -78,13 +81,15 @@ func (mi *ModelsInstall) Run(ctx *cliContext.Context) error {
 			return err
 		}
 
-		model := gallery.FindModel(models, modelName, mi.ModelsPath)
-		if model == nil {
-			log.Error().Str("model", modelName).Msg("model not found")
-			return err
-		}
+		if !downloader.LooksLikeOCI(modelName) {
+			model := gallery.FindModel(models, modelName, mi.ModelsPath)
+			if model == nil {
+				log.Error().Str("model", modelName).Msg("model not found")
+				return err
+			}
 
-		log.Info().Str("model", modelName).Str("license", model.License).Msg("installing model")
+			log.Info().Str("model", modelName).Str("license", model.License).Msg("installing model")
+		}
 		err = startup.InstallModels(galleries, "", mi.ModelsPath, progressCallback, modelName)
 		if err != nil {
 			return err

core/cli/run.go

@@ -3,14 +3,16 @@ package cli
 import (
 	"context"
 	"fmt"
+	"net"
+	"os"
 	"strings"
 	"time"
 
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/http"
-	"github.com/go-skynet/LocalAI/core/p2p"
-	"github.com/go-skynet/LocalAI/core/startup"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http"
+	"github.com/mudler/LocalAI/core/p2p"
+	"github.com/mudler/LocalAI/core/startup"
 	"github.com/rs/zerolog"
 	"github.com/rs/zerolog/log"
 )
@@ -43,13 +45,14 @@ type RunCMD struct {
 	Address              string   `env:"LOCALAI_ADDRESS,ADDRESS" default:":8080" help:"Bind address for the API server" group:"api"`
 	CORS                 bool     `env:"LOCALAI_CORS,CORS" help:"" group:"api"`
 	CORSAllowOrigins     string   `env:"LOCALAI_CORS_ALLOW_ORIGINS,CORS_ALLOW_ORIGINS" group:"api"`
+	LibraryPath          string   `env:"LOCALAI_LIBRARY_PATH,LIBRARY_PATH" help:"Path to the library directory (for e.g. external libraries used by backends)" default:"/usr/share/local-ai/libs" group:"backends"`
 	CSRF                 bool     `env:"LOCALAI_CSRF" help:"Enables fiber CSRF middleware" group:"api"`
 	UploadLimit          int      `env:"LOCALAI_UPLOAD_LIMIT,UPLOAD_LIMIT" default:"15" help:"Default upload-limit in MB" group:"api"`
 	APIKeys              []string `env:"LOCALAI_API_KEY,API_KEY" help:"List of API Keys to enable API authentication. When this is set, all the requests must be authenticated with one of these API keys" group:"api"`
 	DisableWebUI         bool     `env:"LOCALAI_DISABLE_WEBUI,DISABLE_WEBUI" default:"false" help:"Disable webui" group:"api"`
 	OpaqueErrors         bool     `env:"LOCALAI_OPAQUE_ERRORS" default:"false" help:"If true, all error responses are replaced with blank 500 errors. This is intended only for hardening against information leaks and is normally not recommended." group:"api"`
 	Peer2Peer            bool     `env:"LOCALAI_P2P,P2P" name:"p2p" default:"false" help:"Enable P2P mode" group:"p2p"`
-	Peer2PeerToken       string   `env:"LOCALAI_P2P_TOKEN,P2P_TOKEN" name:"p2ptoken" help:"Token for P2P mode (optional)" group:"p2p"`
+	Peer2PeerToken       string   `env:"LOCALAI_P2P_TOKEN,P2P_TOKEN,TOKEN" name:"p2ptoken" help:"Token for P2P mode (optional)" group:"p2p"`
 	ParallelRequests     bool     `env:"LOCALAI_PARALLEL_REQUESTS,PARALLEL_REQUESTS" help:"Enable backends to handle multiple requests in parallel if they support it (e.g.: llama.cpp or vllm)" group:"backends"`
 	SingleActiveBackend  bool     `env:"LOCALAI_SINGLE_ACTIVE_BACKEND,SINGLE_ACTIVE_BACKEND" help:"Allow only one backend to be run at a time" group:"backends"`
 	PreloadBackendOnly   bool     `env:"LOCALAI_PRELOAD_BACKEND_ONLY,PRELOAD_BACKEND_ONLY" default:"false" help:"Do not launch the API services, only the preloaded models / backends are started (useful for multi-node setups)" group:"backends"`
@@ -58,6 +61,7 @@ type RunCMD struct {
 	WatchdogIdleTimeout  string   `env:"LOCALAI_WATCHDOG_IDLE_TIMEOUT,WATCHDOG_IDLE_TIMEOUT" default:"15m" help:"Threshold beyond which an idle backend should be stopped" group:"backends"`
 	EnableWatchdogBusy   bool     `env:"LOCALAI_WATCHDOG_BUSY,WATCHDOG_BUSY" default:"false" help:"Enable watchdog for stopping backends that are busy longer than the watchdog-busy-timeout" group:"backends"`
 	WatchdogBusyTimeout  string   `env:"LOCALAI_WATCHDOG_BUSY_TIMEOUT,WATCHDOG_BUSY_TIMEOUT" default:"5m" help:"Threshold beyond which a busy backend should be stopped" group:"backends"`
+	Federated            bool     `env:"LOCALAI_FEDERATED,FEDERATED" help:"Enable federated instance" group:"federated"`
 }
 
 func (r *RunCMD) Run(ctx *cliContext.Context) error {
@@ -80,6 +84,7 @@ func (r *RunCMD) Run(ctx *cliContext.Context) error {
 		config.WithCors(r.CORS),
 		config.WithCorsAllowOrigins(r.CORSAllowOrigins),
 		config.WithCsrf(r.CSRF),
+		config.WithLibPath(r.LibraryPath),
 		config.WithThreads(r.Threads),
 		config.WithBackendAssets(ctx.BackendAssets),
 		config.WithBackendAssetsOutput(r.BackendAssetsPath),
@@ -89,9 +94,10 @@ func (r *RunCMD) Run(ctx *cliContext.Context) error {
 		config.WithOpaqueErrors(r.OpaqueErrors),
 	}
 
+	token := ""
 	if r.Peer2Peer || r.Peer2PeerToken != "" {
 		log.Info().Msg("P2P mode enabled")
-		token := r.Peer2PeerToken
+		token = r.Peer2PeerToken
 		if token == "" {
 			// IF no token is provided, and p2p is enabled,
 			// we generate one and wait for the user to pick up the token (this is for interactive)
@@ -102,14 +108,46 @@ func (r *RunCMD) Run(ctx *cliContext.Context) error {
 
 			log.Info().Msg("To use the token, you can run the following command in another node or terminal:")
 			fmt.Printf("export TOKEN=\"%s\"\nlocal-ai worker p2p-llama-cpp-rpc\n", token)
-
-			// Ask for user confirmation
-			log.Info().Msg("Press a button to proceed")
-			var input string
-			fmt.Scanln(&input)
 		}
+		opts = append(opts, config.WithP2PToken(token))
+
+		node, err := p2p.NewNode(token)
+		if err != nil {
+			return err
+		}
+
 		log.Info().Msg("Starting P2P server discovery...")
-		if err := p2p.LLamaCPPRPCServerDiscoverer(context.Background(), token); err != nil {
+		if err := p2p.ServiceDiscoverer(context.Background(), node, token, "", func() {
+			var tunnelAddresses []string
+			for _, v := range p2p.GetAvailableNodes("") {
+				if v.IsOnline() {
+					tunnelAddresses = append(tunnelAddresses, v.TunnelAddress)
+				} else {
+					log.Info().Msgf("Node %s is offline", v.ID)
+				}
+			}
+			tunnelEnvVar := strings.Join(tunnelAddresses, ",")
+
+			os.Setenv("LLAMACPP_GRPC_SERVERS", tunnelEnvVar)
+			log.Debug().Msgf("setting LLAMACPP_GRPC_SERVERS to %s", tunnelEnvVar)
+		}); err != nil {
+			return err
+		}
+	}
+
+	if r.Federated {
+		_, port, err := net.SplitHostPort(r.Address)
+		if err != nil {
+			return err
+		}
+		if err := p2p.ExposeService(context.Background(), "localhost", port, token, p2p.FederatedID); err != nil {
+			return err
+		}
+		node, err := p2p.NewNode(token)
+		if err != nil {
+			return err
+		}
+		if err := p2p.ServiceDiscoverer(context.Background(), node, token, p2p.FederatedID, nil); err != nil {
 			return err
 		}
 	}

core/cli/transcript.go

@@ -5,10 +5,10 @@ import (
 	"errors"
 	"fmt"
 
-	"github.com/go-skynet/LocalAI/core/backend"
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/backend"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
 
@@ -18,6 +18,7 @@ type TranscriptCMD struct {
 	Backend           string `short:"b" default:"whisper" help:"Backend to run the transcription model"`
 	Model             string `short:"m" required:"" help:"Model name to run the TTS"`
 	Language          string `short:"l" help:"Language of the audio file"`
+	Translate         bool   `short:"c" help:"Translate the transcription to english"`
 	Threads           int    `short:"t" default:"1" help:"Number of threads used for parallel computation"`
 	ModelsPath        string `env:"LOCALAI_MODELS_PATH,MODELS_PATH" type:"path" default:"${basepath}/models" help:"Path containing models used for inferencing" group:"storage"`
 	BackendAssetsPath string `env:"LOCALAI_BACKEND_ASSETS_PATH,BACKEND_ASSETS_PATH" type:"path" default:"/tmp/localai/backend_data" help:"Path used to extract libraries that are required by some of the backends in runtime" group:"storage"`
@@ -50,7 +51,7 @@ func (t *TranscriptCMD) Run(ctx *cliContext.Context) error {
 		}
 	}()
 
-	tr, err := backend.ModelTranscription(t.Filename, t.Language, ml, c, opts)
+	tr, err := backend.ModelTranscription(t.Filename, t.Language, t.Translate, ml, c, opts)
 	if err != nil {
 		return err
 	}

core/cli/tts.go

@@ -7,10 +7,10 @@ import (
 	"path/filepath"
 	"strings"
 
-	"github.com/go-skynet/LocalAI/core/backend"
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/backend"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
 

core/cli/util.go

@@ -5,7 +5,7 @@ import (
 
 	"github.com/rs/zerolog/log"
 
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
 	gguf "github.com/thxcode/gguf-parser-go"
 )
 

core/cli/worker/worker_llamacpp.go

@@ -5,8 +5,9 @@ import (
 	"os"
 	"syscall"
 
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
-	"github.com/go-skynet/LocalAI/pkg/assets"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/pkg/assets"
+	"github.com/mudler/LocalAI/pkg/library"
 	"github.com/rs/zerolog/log"
 )
 
@@ -27,17 +28,18 @@ func (r *LLamaCPP) Run(ctx *cliContext.Context) error {
 		return fmt.Errorf("usage: local-ai worker llama-cpp-rpc -- <llama-rpc-server-args>")
 	}
 
+	grpcProcess := assets.ResolvePath(
+		r.BackendAssetsPath,
+		"util",
+		"llama-cpp-rpc-server",
+	)
+
+	args := os.Args[4:]
+	args, grpcProcess = library.LoadLDSO(r.BackendAssetsPath, args, grpcProcess)
+
+	args = append([]string{grpcProcess}, args...)
 	return syscall.Exec(
-		assets.ResolvePath(
-			r.BackendAssetsPath,
-			"util",
-			"llama-cpp-rpc-server",
-		),
-		append([]string{
-			assets.ResolvePath(
-				r.BackendAssetsPath,
-				"util",
-				"llama-cpp-rpc-server",
-			)}, os.Args[4:]...),
+		grpcProcess,
+		args,
 		os.Environ())
 }

core/cli/worker/worker_nop2p.go

@@ -6,7 +6,7 @@ package worker
 import (
 	"fmt"
 
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
 )
 
 type P2P struct{}

core/cli/worker/worker_p2p.go

@@ -10,16 +10,17 @@ import (
 	"os/exec"
 	"time"
 
-	cliContext "github.com/go-skynet/LocalAI/core/cli/context"
-	"github.com/go-skynet/LocalAI/core/p2p"
-	"github.com/go-skynet/LocalAI/pkg/assets"
+	cliContext "github.com/mudler/LocalAI/core/cli/context"
+	"github.com/mudler/LocalAI/core/p2p"
+	"github.com/mudler/LocalAI/pkg/assets"
+	"github.com/mudler/LocalAI/pkg/library"
 	"github.com/phayes/freeport"
 	"github.com/rs/zerolog/log"
 )
 
 type P2P struct {
 	WorkerFlags       `embed:""`
-	Token             string   `env:"LOCALAI_TOKEN,TOKEN" help:"JSON list of galleries"`
+	Token             string   `env:"LOCALAI_TOKEN,LOCALAI_P2P_TOKEN,TOKEN" help:"P2P token to use"`
 	NoRunner          bool     `env:"LOCALAI_NO_RUNNER,NO_RUNNER" help:"Do not start the llama-cpp-rpc-server"`
 	RunnerAddress     string   `env:"LOCALAI_RUNNER_ADDRESS,RUNNER_ADDRESS" help:"Address of the llama-cpp-rpc-server"`
 	RunnerPort        string   `env:"LOCALAI_RUNNER_PORT,RUNNER_PORT" help:"Port of the llama-cpp-rpc-server"`
@@ -58,7 +59,7 @@ func (r *P2P) Run(ctx *cliContext.Context) error {
 			p = r.RunnerPort
 		}
 
-		err = p2p.BindLLamaCPPWorker(context.Background(), address, p, r.Token)
+		err = p2p.ExposeService(context.Background(), address, p, r.Token, "")
 		if err != nil {
 			return err
 		}
@@ -71,13 +72,18 @@ func (r *P2P) Run(ctx *cliContext.Context) error {
 	go func() {
 		for {
 			log.Info().Msgf("Starting llama-cpp-rpc-server on '%s:%d'", address, port)
+
+			grpcProcess := assets.ResolvePath(
+				r.BackendAssetsPath,
+				"util",
+				"llama-cpp-rpc-server",
+			)
+
+			args := append([]string{"--host", address, "--port", fmt.Sprint(port)}, r.ExtraLLamaCPPArgs...)
+			args, grpcProcess = library.LoadLDSO(r.BackendAssetsPath, args, grpcProcess)
+
 			cmd := exec.Command(
-				assets.ResolvePath(
-					r.BackendAssetsPath,
-					"util",
-					"llama-cpp-rpc-server",
-				),
-				append([]string{"--host", address, "--port", fmt.Sprint(port)}, r.ExtraLLamaCPPArgs...)...,
+				grpcProcess, args...,
 			)
 
 			cmd.Env = os.Environ()
@@ -86,14 +92,14 @@ func (r *P2P) Run(ctx *cliContext.Context) error {
 			cmd.Stdout = os.Stdout
 
 			if err := cmd.Start(); err != nil {
-				log.Error().Err(err).Msg("Failed to start llama-cpp-rpc-server")
+				log.Error().Any("grpcProcess", grpcProcess).Any("args", args).Err(err).Msg("Failed to start llama-cpp-rpc-server")
 			}
 
 			cmd.Wait()
 		}
 	}()
 
-	err = p2p.BindLLamaCPPWorker(context.Background(), address, fmt.Sprint(port), r.Token)
+	err = p2p.ExposeService(context.Background(), address, fmt.Sprint(port), r.Token, "")
 	if err != nil {
 		return err
 	}

core/config/application_config.go

@@ -6,8 +6,7 @@ import (
 	"encoding/json"
 	"time"
 
-	"github.com/go-skynet/LocalAI/pkg/gallery"
-	"github.com/go-skynet/LocalAI/pkg/xsysinfo"
+	"github.com/mudler/LocalAI/pkg/xsysinfo"
 	"github.com/rs/zerolog/log"
 )
 
@@ -15,6 +14,7 @@ type ApplicationConfig struct {
 	Context                             context.Context
 	ConfigFile                          string
 	ModelPath                           string
+	LibPath                             string
 	UploadLimitMB, Threads, ContextSize int
 	DisableWebUI                        bool
 	F16                                 bool
@@ -32,10 +32,11 @@ type ApplicationConfig struct {
 	CORSAllowOrigins                    string
 	ApiKeys                             []string
 	OpaqueErrors                        bool
+	P2PToken                            string
 
 	ModelLibraryURL string
 
-	Galleries []gallery.Gallery
+	Galleries []Gallery
 
 	BackendAssets     embed.FS
 	AssetsDestination string
@@ -95,12 +96,24 @@ func WithCsrf(b bool) AppOption {
 	}
 }
 
+func WithP2PToken(s string) AppOption {
+	return func(o *ApplicationConfig) {
+		o.P2PToken = s
+	}
+}
+
 func WithModelLibraryURL(url string) AppOption {
 	return func(o *ApplicationConfig) {
 		o.ModelLibraryURL = url
 	}
 }
 
+func WithLibPath(path string) AppOption {
+	return func(o *ApplicationConfig) {
+		o.LibPath = path
+	}
+}
+
 var EnableWatchDog = func(o *ApplicationConfig) {
 	o.WatchDog = true
 }
@@ -173,10 +186,10 @@ func WithBackendAssets(f embed.FS) AppOption {
 func WithStringGalleries(galls string) AppOption {
 	return func(o *ApplicationConfig) {
 		if galls == "" {
-			o.Galleries = []gallery.Gallery{}
+			o.Galleries = []Gallery{}
 			return
 		}
-		var galleries []gallery.Gallery
+		var galleries []Gallery
 		if err := json.Unmarshal([]byte(galls), &galleries); err != nil {
 			log.Error().Err(err).Msg("failed loading galleries")
 		}
@@ -184,7 +197,7 @@ func WithStringGalleries(galls string) AppOption {
 	}
 }
 
-func WithGalleries(galleries []gallery.Gallery) AppOption {
+func WithGalleries(galleries []Gallery) AppOption {
 	return func(o *ApplicationConfig) {
 		o.Galleries = append(o.Galleries, galleries...)
 	}

core/config/backend_config.go

@@ -5,10 +5,10 @@ import (
 	"regexp"
 	"strings"
 
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/downloader"
-	"github.com/go-skynet/LocalAI/pkg/functions"
-	"github.com/go-skynet/LocalAI/pkg/utils"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/downloader"
+	"github.com/mudler/LocalAI/pkg/functions"
+	"github.com/mudler/LocalAI/pkg/utils"
 )
 
 const (
@@ -390,10 +390,6 @@ func (c *BackendConfig) Validate() bool {
 		}
 	}
 
-	if c.Name == "" {
-		return false
-	}
-
 	if c.Backend != "" {
 		// a regex that checks that is a string name with no special characters, except '-' and '_'
 		re := regexp.MustCompile(`^[a-zA-Z0-9-_]+$`)

core/config/backend_config_loader.go

@@ -11,9 +11,9 @@ import (
 	"sync"
 
 	"github.com/charmbracelet/glamour"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/downloader"
-	"github.com/go-skynet/LocalAI/pkg/utils"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/downloader"
+	"github.com/mudler/LocalAI/pkg/utils"
 	"github.com/rs/zerolog/log"
 	"gopkg.in/yaml.v3"
 )

core/config/backend_config_test.go

@@ -16,7 +16,8 @@ var _ = Describe("Test cases for config related functions", func() {
 			Expect(err).To(BeNil())
 			defer os.Remove(tmp.Name())
 			_, err = tmp.WriteString(
-				`backend: "foo-bar"
+				`backend: "../foo-bar"
+name: "foo"
 parameters:
   model: "foo-bar"`)
 			Expect(err).ToNot(HaveOccurred())

core/config/gallery.go

@@ -0,0 +1,6 @@
+package config
+
+type Gallery struct {
+	URL  string `json:"url" yaml:"url"`
+	Name string `json:"name" yaml:"name"`
+}

core/dependencies_manager/manager.go

@@ -5,8 +5,8 @@ import (
 	"os"
 	"path/filepath"
 
-	"github.com/go-skynet/LocalAI/pkg/downloader"
-	"github.com/go-skynet/LocalAI/pkg/utils"
+	"github.com/mudler/LocalAI/pkg/downloader"
+	"github.com/mudler/LocalAI/pkg/utils"
 	"gopkg.in/yaml.v3"
 )
 

core/gallery/gallery.go

@@ -7,19 +7,15 @@ import (
 	"path/filepath"
 	"strings"
 
-	"github.com/go-skynet/LocalAI/pkg/downloader"
 	"github.com/imdario/mergo"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/pkg/downloader"
 	"github.com/rs/zerolog/log"
 	"gopkg.in/yaml.v2"
 )
 
-type Gallery struct {
-	URL  string `json:"url" yaml:"url"`
-	Name string `json:"name" yaml:"name"`
-}
-
 // Installs a model from the gallery
-func InstallModelFromGallery(galleries []Gallery, name string, basePath string, req GalleryModel, downloadStatus func(string, string, string, float64)) error {
+func InstallModelFromGallery(galleries []config.Gallery, name string, basePath string, req GalleryModel, downloadStatus func(string, string, string, float64)) error {
 
 	applyModel := func(model *GalleryModel) error {
 		name = strings.ReplaceAll(name, string(os.PathSeparator), "__")
@@ -117,7 +113,7 @@ func FindModel(models []*GalleryModel, name string, basePath string) *GalleryMod
 // List available models
 // Models galleries are a list of yaml files that are hosted on a remote server (for example github).
 // Each yaml file contains a list of models that can be downloaded and optionally overrides to define a new model setting.
-func AvailableGalleryModels(galleries []Gallery, basePath string) ([]*GalleryModel, error) {
+func AvailableGalleryModels(galleries []config.Gallery, basePath string) ([]*GalleryModel, error) {
 	var models []*GalleryModel
 
 	// Get models from galleries
@@ -134,7 +130,7 @@ func AvailableGalleryModels(galleries []Gallery, basePath string) ([]*GalleryMod
 
 func findGalleryURLFromReferenceURL(url string, basePath string) (string, error) {
 	var refFile string
-	err := downloader.GetURI(url, basePath, func(url string, d []byte) error {
+	err := downloader.DownloadAndUnmarshal(url, basePath, func(url string, d []byte) error {
 		refFile = string(d)
 		if len(refFile) == 0 {
 			return fmt.Errorf("invalid reference file at url %s: %s", url, d)
@@ -146,7 +142,7 @@ func findGalleryURLFromReferenceURL(url string, basePath string) (string, error)
 	return refFile, err
 }
 
-func getGalleryModels(gallery Gallery, basePath string) ([]*GalleryModel, error) {
+func getGalleryModels(gallery config.Gallery, basePath string) ([]*GalleryModel, error) {
 	var models []*GalleryModel = []*GalleryModel{}
 
 	if strings.HasSuffix(gallery.URL, ".ref") {
@@ -157,7 +153,7 @@ func getGalleryModels(gallery Gallery, basePath string) ([]*GalleryModel, error)
 		}
 	}
 
-	err := downloader.GetURI(gallery.URL, basePath, func(url string, d []byte) error {
+	err := downloader.DownloadAndUnmarshal(gallery.URL, basePath, func(url string, d []byte) error {
 		return yaml.Unmarshal(d, &models)
 	})
 	if err != nil {

core/gallery/models.go

@@ -5,9 +5,11 @@ import (
 	"os"
 	"path/filepath"
 
-	"github.com/go-skynet/LocalAI/pkg/downloader"
-	"github.com/go-skynet/LocalAI/pkg/utils"
 	"github.com/imdario/mergo"
+	lconfig "github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/pkg/downloader"
+	"github.com/mudler/LocalAI/pkg/utils"
+
 	"github.com/rs/zerolog/log"
 	"gopkg.in/yaml.v2"
 )
@@ -65,7 +67,7 @@ type PromptTemplate struct {
 
 func GetGalleryConfigFromURL(url string, basePath string) (Config, error) {
 	var config Config
-	err := downloader.GetURI(url, basePath, func(url string, d []byte) error {
+	err := downloader.DownloadAndUnmarshal(url, basePath, func(url string, d []byte) error {
 		return yaml.Unmarshal(d, &config)
 	})
 	if err != nil {
@@ -172,6 +174,15 @@ func InstallModel(basePath, nameOverride string, config *Config, configOverrides
 			return fmt.Errorf("failed to marshal updated config YAML: %v", err)
 		}
 
+		backendConfig := lconfig.BackendConfig{}
+		err = yaml.Unmarshal(updatedConfigYAML, &backendConfig)
+		if err != nil {
+			return fmt.Errorf("failed to unmarshal updated config YAML: %v", err)
+		}
+		if !backendConfig.Validate() {
+			return fmt.Errorf("failed to validate updated config YAML")
+		}
+
 		err = os.WriteFile(configFilePath, updatedConfigYAML, 0600)
 		if err != nil {
 			return fmt.Errorf("failed to write updated config file: %v", err)

core/gallery/models_test.go

@@ -5,7 +5,8 @@ import (
 	"os"
 	"path/filepath"
 
-	. "github.com/go-skynet/LocalAI/pkg/gallery"
+	"github.com/mudler/LocalAI/core/config"
+	. "github.com/mudler/LocalAI/core/gallery"
 	. "github.com/onsi/ginkgo/v2"
 	. "github.com/onsi/gomega"
 	"gopkg.in/yaml.v3"
@@ -54,7 +55,7 @@ var _ = Describe("Model test", func() {
 			err = os.WriteFile(galleryFilePath, out, 0600)
 			Expect(err).ToNot(HaveOccurred())
 			Expect(filepath.IsAbs(galleryFilePath)).To(BeTrue(), galleryFilePath)
-			galleries := []Gallery{
+			galleries := []config.Gallery{
 				{
 					Name: "test",
 					URL:  "file://" + galleryFilePath,

core/gallery/op.go

@@ -1,5 +1,7 @@
 package gallery
 
+import "github.com/mudler/LocalAI/core/config"
+
 type GalleryOp struct {
 	Id               string
 	GalleryModelName string
@@ -7,7 +9,7 @@ type GalleryOp struct {
 	Delete           bool
 
 	Req       GalleryModel
-	Galleries []Gallery
+	Galleries []config.Gallery
 }
 
 type GalleryOpStatus struct {

core/gallery/request.go

@@ -3,6 +3,8 @@ package gallery
 import (
 	"fmt"
 	"strings"
+
+	"github.com/mudler/LocalAI/core/config"
 )
 
 // GalleryModel is the struct used to represent a model in the gallery returned by the endpoint.
@@ -23,7 +25,7 @@ type GalleryModel struct {
 	// AdditionalFiles are used to add additional files to the model
 	AdditionalFiles []File `json:"files,omitempty" yaml:"files,omitempty"`
 	// Gallery is a reference to the gallery which contains the model
-	Gallery Gallery `json:"gallery,omitempty" yaml:"gallery,omitempty"`
+	Gallery config.Gallery `json:"gallery,omitempty" yaml:"gallery,omitempty"`
 	// Installed is used to indicate if the model is installed or not
 	Installed bool `json:"installed,omitempty" yaml:"installed,omitempty"`
 }

core/gallery/request_test.go

@@ -1,7 +1,7 @@
 package gallery_test
 
 import (
-	. "github.com/go-skynet/LocalAI/pkg/gallery"
+	. "github.com/mudler/LocalAI/core/gallery"
 	. "github.com/onsi/ginkgo/v2"
 	. "github.com/onsi/gomega"
 )

core/http/app.go

@@ -6,16 +6,16 @@ import (
 	"net/http"
 	"strings"
 
-	"github.com/go-skynet/LocalAI/pkg/utils"
+	"github.com/mudler/LocalAI/pkg/utils"
 
-	"github.com/go-skynet/LocalAI/core/http/endpoints/localai"
-	"github.com/go-skynet/LocalAI/core/http/endpoints/openai"
-	"github.com/go-skynet/LocalAI/core/http/routes"
+	"github.com/mudler/LocalAI/core/http/endpoints/localai"
+	"github.com/mudler/LocalAI/core/http/endpoints/openai"
+	"github.com/mudler/LocalAI/core/http/routes"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/core/services"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/services"
+	"github.com/mudler/LocalAI/pkg/model"
 
 	"github.com/gofiber/contrib/fiberzerolog"
 	"github.com/gofiber/fiber/v2"

core/http/app_test.go

@@ -13,15 +13,15 @@ import (
 	"path/filepath"
 	"runtime"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	. "github.com/go-skynet/LocalAI/core/http"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/core/startup"
+	"github.com/mudler/LocalAI/core/config"
+	. "github.com/mudler/LocalAI/core/http"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/startup"
 
-	"github.com/go-skynet/LocalAI/pkg/downloader"
-	"github.com/go-skynet/LocalAI/pkg/gallery"
-	"github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/pkg/downloader"
+	"github.com/mudler/LocalAI/pkg/model"
 	. "github.com/onsi/ginkgo/v2"
 	. "github.com/onsi/gomega"
 	"gopkg.in/yaml.v3"
@@ -74,7 +74,7 @@ func getModelStatus(url string) (response map[string]interface{}) {
 
 func getModels(url string) (response []gallery.GalleryModel) {
 	// TODO: No tests currently seem to exercise file:// urls. Fix?
-	downloader.GetURI(url, "", func(url string, i []byte) error {
+	downloader.DownloadAndUnmarshal(url, "", func(url string, i []byte) error {
 		// Unmarshal YAML data into a struct
 		return json.Unmarshal(i, &response)
 	})
@@ -247,7 +247,7 @@ var _ = Describe("API test", func() {
 			err = os.WriteFile(filepath.Join(modelDir, "gallery_simple.yaml"), out, 0600)
 			Expect(err).ToNot(HaveOccurred())
 
-			galleries := []gallery.Gallery{
+			galleries := []config.Gallery{
 				{
 					Name: "test",
 					URL:  "file://" + filepath.Join(modelDir, "gallery_simple.yaml"),
@@ -603,7 +603,7 @@ var _ = Describe("API test", func() {
 
 			c, cancel = context.WithCancel(context.Background())
 
-			galleries := []gallery.Gallery{
+			galleries := []config.Gallery{
 				{
 					Name: "model-gallery",
 					URL:  "https://raw.githubusercontent.com/go-skynet/model-gallery/main/index.yaml",

core/http/ctx/fiber.go

@@ -4,8 +4,8 @@ import (
 	"fmt"
 	"strings"
 
-	"github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
 

core/http/elements/gallery.go

@@ -6,15 +6,24 @@ import (
 
 	"github.com/chasefleming/elem-go"
 	"github.com/chasefleming/elem-go/attrs"
-	"github.com/go-skynet/LocalAI/core/services"
-	"github.com/go-skynet/LocalAI/pkg/gallery"
-	"github.com/go-skynet/LocalAI/pkg/xsync"
+	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/core/p2p"
+	"github.com/mudler/LocalAI/core/services"
+	"github.com/mudler/LocalAI/pkg/xsync"
 )
 
 const (
 	noImage = "https://upload.wikimedia.org/wikipedia/commons/6/65/No-Image-Placeholder.svg"
 )
 
+func renderElements(n []elem.Node) string {
+	render := ""
+	for _, r := range n {
+		render += r.Render()
+	}
+	return render
+}
+
 func DoneProgress(galleryID, text string, showDelete bool) string {
 	var modelName = galleryID
 	// Split by @ and grab the name
@@ -72,6 +81,135 @@ func ProgressBar(progress string) string {
 	).Render()
 }
 
+func P2PNodeStats(nodes []p2p.NodeData) string {
+	/*
+	   <div class="bg-gray-800 p-6 rounded-lg shadow-lg text-left">
+	                       <p class="text-xl font-semibold text-gray-200">Total Workers Detected: {{ len .Nodes }}</p>
+	                       {{ $online := 0 }}
+	                       {{ range .Nodes }}
+	                           {{ if .IsOnline }}
+	                               {{ $online = add $online 1 }}
+	                           {{ end }}
+	                       {{ end }}
+	                       <p class="text-xl font-semibold text-gray-200">Total Online Workers: {{$online}}</p>
+	                   </div>
+	*/
+
+	online := 0
+	for _, n := range nodes {
+		if n.IsOnline() {
+			online++
+		}
+	}
+
+	class := "text-green-500"
+	if online == 0 {
+		class = "text-red-500"
+	}
+	/*
+	   <i class="fas fa-circle animate-pulse text-green-500 ml-2 mr-1"></i>
+	*/
+	circle := elem.I(attrs.Props{
+		"class": "fas fa-circle animate-pulse " + class + " ml-2 mr-1",
+	})
+	nodesElements := []elem.Node{
+		elem.Span(
+			attrs.Props{
+				"class": class,
+			},
+			circle,
+			elem.Text(fmt.Sprintf("%d", online)),
+		),
+		elem.Span(
+			attrs.Props{
+				"class": "text-gray-200",
+			},
+			elem.Text(fmt.Sprintf("/%d", len(nodes))),
+		),
+	}
+
+	return renderElements(nodesElements)
+}
+
+func P2PNodeBoxes(nodes []p2p.NodeData) string {
+	/*
+			<div class="bg-gray-800 p-4 rounded-lg shadow-lg text-left">
+			<div class="flex items-center mb-2">
+				<i class="fas fa-desktop text-gray-400 mr-2"></i>
+				<span class="text-gray-200 font-semibold">{{.ID}}</span>
+			</div>
+			<p class="text-sm text-gray-400 mt-2 flex items-center">
+				Status:
+				<i class="fas fa-circle {{ if .IsOnline }}text-green-500{{ else }}text-red-500{{ end }} ml-2 mr-1"></i>
+				<span class="{{ if .IsOnline }}text-green-400{{ else }}text-red-400{{ end }}">
+					{{ if .IsOnline }}Online{{ else }}Offline{{ end }}
+				</span>
+			</p>
+		</div>
+	*/
+
+	nodesElements := []elem.Node{}
+
+	for _, n := range nodes {
+
+		nodesElements = append(nodesElements,
+			elem.Div(
+				attrs.Props{
+					"class": "bg-gray-700 p-6 rounded-lg shadow-lg text-left",
+				},
+				elem.P(
+					attrs.Props{
+						"class": "text-sm text-gray-400 mt-2 flex",
+					},
+					elem.I(
+						attrs.Props{
+							"class": "fas fa-desktop text-gray-400 mr-2",
+						},
+					),
+					elem.Text("Name: "),
+					elem.Span(
+						attrs.Props{
+							"class": "text-gray-200 font-semibold ml-2 mr-1",
+						},
+						elem.Text(n.ID),
+					),
+					elem.Text("Status: "),
+					elem.If(
+						n.IsOnline(),
+						elem.I(
+							attrs.Props{
+								"class": "fas fa-circle animate-pulse text-green-500 ml-2 mr-1",
+							},
+						),
+						elem.I(
+							attrs.Props{
+								"class": "fas fa-circle animate-pulse text-red-500 ml-2 mr-1",
+							},
+						),
+					),
+					elem.If(
+						n.IsOnline(),
+						elem.Span(
+							attrs.Props{
+								"class": "text-green-400",
+							},
+
+							elem.Text("Online"),
+						),
+						elem.Span(
+							attrs.Props{
+								"class": "text-red-400",
+							},
+							elem.Text("Offline"),
+						),
+					),
+				),
+			))
+	}
+
+	return renderElements(nodesElements)
+}
+
 func StartProgressBar(uid, progress, text string) string {
 	if progress == "" {
 		progress = "0"

core/http/endpoints/elevenlabs/tts.go

@@ -1,13 +1,13 @@
 package elevenlabs
 
 import (
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
-	fiberContext "github.com/go-skynet/LocalAI/core/http/ctx"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
+	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/pkg/model"
 
-	"github.com/go-skynet/LocalAI/core/schema"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/rs/zerolog/log"
 )
 

core/http/endpoints/jina/rerank.go

@@ -1,14 +1,14 @@
 package jina
 
 import (
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
 
-	fiberContext "github.com/go-skynet/LocalAI/core/http/ctx"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/grpc/proto"
-	"github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
+	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
 

core/http/endpoints/localai/backend_monitor.go

@@ -1,9 +1,9 @@
 package localai
 
 import (
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/core/services"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/services"
 )
 
 func BackendMonitorEndpoint(bm *services.BackendMonitorService) func(c *fiber.Ctx) error {

core/http/endpoints/localai/gallery.go

@@ -5,15 +5,16 @@ import (
 	"fmt"
 	"slices"
 
-	"github.com/go-skynet/LocalAI/core/services"
-	"github.com/go-skynet/LocalAI/pkg/gallery"
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/core/services"
 	"github.com/rs/zerolog/log"
 )
 
 type ModelGalleryEndpointService struct {
-	galleries      []gallery.Gallery
+	galleries      []config.Gallery
 	modelPath      string
 	galleryApplier *services.GalleryService
 }
@@ -24,7 +25,7 @@ type GalleryModel struct {
 	gallery.GalleryModel
 }
 
-func CreateModelGalleryEndpointService(galleries []gallery.Gallery, modelPath string, galleryApplier *services.GalleryService) ModelGalleryEndpointService {
+func CreateModelGalleryEndpointService(galleries []config.Gallery, modelPath string, galleryApplier *services.GalleryService) ModelGalleryEndpointService {
 	return ModelGalleryEndpointService{
 		galleries:      galleries,
 		modelPath:      modelPath,
@@ -129,12 +130,12 @@ func (mgs *ModelGalleryEndpointService) ListModelGalleriesEndpoint() func(c *fib
 
 func (mgs *ModelGalleryEndpointService) AddModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		input := new(gallery.Gallery)
+		input := new(config.Gallery)
 		// Get input data from the request body
 		if err := c.BodyParser(input); err != nil {
 			return err
 		}
-		if slices.ContainsFunc(mgs.galleries, func(gallery gallery.Gallery) bool {
+		if slices.ContainsFunc(mgs.galleries, func(gallery config.Gallery) bool {
 			return gallery.Name == input.Name
 		}) {
 			return fmt.Errorf("%s already exists", input.Name)
@@ -151,17 +152,17 @@ func (mgs *ModelGalleryEndpointService) AddModelGalleryEndpoint() func(c *fiber.
 
 func (mgs *ModelGalleryEndpointService) RemoveModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		input := new(gallery.Gallery)
+		input := new(config.Gallery)
 		// Get input data from the request body
 		if err := c.BodyParser(input); err != nil {
 			return err
 		}
-		if !slices.ContainsFunc(mgs.galleries, func(gallery gallery.Gallery) bool {
+		if !slices.ContainsFunc(mgs.galleries, func(gallery config.Gallery) bool {
 			return gallery.Name == input.Name
 		}) {
 			return fmt.Errorf("%s is not currently registered", input.Name)
 		}
-		mgs.galleries = slices.DeleteFunc(mgs.galleries, func(gallery gallery.Gallery) bool {
+		mgs.galleries = slices.DeleteFunc(mgs.galleries, func(gallery config.Gallery) bool {
 			return gallery.Name == input.Name
 		})
 		return c.Send(nil)

core/http/endpoints/localai/metrics.go

@@ -3,9 +3,9 @@ package localai
 import (
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/services"
 	"github.com/gofiber/fiber/v2"
 	"github.com/gofiber/fiber/v2/middleware/adaptor"
+	"github.com/mudler/LocalAI/core/services"
 	"github.com/prometheus/client_golang/prometheus/promhttp"
 )
 

core/http/endpoints/localai/stores.go

@@ -1,12 +1,12 @@
 package localai
 
 import (
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/go-skynet/LocalAI/pkg/store"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/store"
 )
 
 func StoresSetEndpoint(sl *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {

core/http/endpoints/localai/tts.go

@@ -1,13 +1,13 @@
 package localai
 
 import (
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
-	fiberContext "github.com/go-skynet/LocalAI/core/http/ctx"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
+	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/pkg/model"
 
-	"github.com/go-skynet/LocalAI/core/schema"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/rs/zerolog/log"
 )
 

core/http/endpoints/localai/welcome.go

@@ -1,11 +1,12 @@
 package localai
 
 import (
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/internal"
-	"github.com/go-skynet/LocalAI/pkg/gallery"
-	"github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/core/p2p"
+	"github.com/mudler/LocalAI/internal"
+	"github.com/mudler/LocalAI/pkg/model"
 )
 
 func WelcomeEndpoint(appConfig *config.ApplicationConfig,
@@ -33,6 +34,7 @@ func WelcomeEndpoint(appConfig *config.ApplicationConfig,
 			"Models":            models,
 			"ModelsConfig":      backendConfigs,
 			"GalleryConfig":     galleryConfigs,
+			"IsP2PEnabled":      p2p.IsP2PEnabled(),
 			"ApplicationConfig": appConfig,
 			"ProcessingModels":  processingModels,
 			"TaskTypes":         taskTypes,

core/http/endpoints/openai/assistant.go

@@ -9,10 +9,10 @@ import (
 	"sync/atomic"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	model "github.com/go-skynet/LocalAI/pkg/model"
-	"github.com/go-skynet/LocalAI/pkg/utils"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/config"
+	model "github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/utils"
 	"github.com/rs/zerolog/log"
 )
 
@@ -339,7 +339,7 @@ func CreateAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 			}
 		}
 
-		return c.Status(fiber.StatusNotFound).SendString(fmt.Sprintf("Unable to find "))
+		return c.Status(fiber.StatusNotFound).SendString(fmt.Sprintf("Unable to find %q", assistantID))
 	}
 }
 

core/http/endpoints/openai/assistant_test.go

@@ -4,7 +4,6 @@ import (
 	"encoding/json"
 	"fmt"
 	"io"
-	"io/ioutil"
 	"net/http"
 	"net/http/httptest"
 	"os"
@@ -13,9 +12,9 @@ import (
 	"testing"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/stretchr/testify/assert"
 )
 
@@ -183,7 +182,7 @@ func TestAssistantEndpoints(t *testing.T) {
 				assert.NoError(t, err)
 				assert.Equal(t, tt.expectedStatus, response.StatusCode)
 				if tt.expectedStatus != fiber.StatusOK {
-					all, _ := ioutil.ReadAll(response.Body)
+					all, _ := io.ReadAll(response.Body)
 					assert.Equal(t, tt.expectedStringResult, string(all))
 				} else {
 					var result []Assistant
@@ -279,6 +278,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		assert.NoError(t, err)
 		var getAssistant Assistant
 		err = json.NewDecoder(modifyResponse.Body).Decode(&getAssistant)
+		assert.NoError(t, err)
 
 		t.Cleanup(cleanupAllAssistants(t, app, []string{getAssistant.ID}))
 
@@ -391,7 +391,10 @@ func createAssistantFile(app *fiber.App, afr AssistantFileRequest, assistantId s
 	}
 
 	var assistantFile AssistantFile
-	all, err := ioutil.ReadAll(resp.Body)
+	all, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return AssistantFile{}, resp, err
+	}
 	err = json.NewDecoder(strings.NewReader(string(all))).Decode(&assistantFile)
 	if err != nil {
 		return AssistantFile{}, resp, err
@@ -422,8 +425,7 @@ func createAssistant(app *fiber.App, ar AssistantRequest) (Assistant, *http.Resp
 
 	var resultAssistant Assistant
 	err = json.NewDecoder(strings.NewReader(string(bodyString))).Decode(&resultAssistant)
-
-	return resultAssistant, resp, nil
+	return resultAssistant, resp, err
 }
 
 func cleanupAllAssistants(t *testing.T, app *fiber.App, ids []string) func() {

core/http/endpoints/openai/chat.go

@@ -8,13 +8,13 @@ import (
 	"strings"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/functions"
-	model "github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/functions"
+	model "github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 	"github.com/valyala/fasthttp"
 )

core/http/endpoints/openai/completion.go

@@ -8,14 +8,14 @@ import (
 	"fmt"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/functions"
-	model "github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/functions"
+	model "github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 	"github.com/valyala/fasthttp"
 )

core/http/endpoints/openai/edit.go

@@ -5,13 +5,13 @@ import (
 	"fmt"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/core/schema"
-	model "github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
+	"github.com/mudler/LocalAI/core/schema"
+	model "github.com/mudler/LocalAI/pkg/model"
 
 	"github.com/rs/zerolog/log"
 )

core/http/endpoints/openai/embeddings.go

@@ -5,12 +5,12 @@ import (
 	"fmt"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/pkg/model"
 
-	"github.com/go-skynet/LocalAI/core/schema"
 	"github.com/google/uuid"
+	"github.com/mudler/LocalAI/core/schema"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/rs/zerolog/log"

core/http/endpoints/openai/files.go

@@ -8,10 +8,10 @@ import (
 	"sync/atomic"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/pkg/utils"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/pkg/utils"
 )
 
 var UploadedFiles []File
@@ -123,7 +123,10 @@ func getFileFromRequest(c *fiber.Ctx) (*File, error) {
 	return nil, fmt.Errorf("unable to find file id %s", id)
 }
 
-// GetFilesEndpoint https://platform.openai.com/docs/api-reference/files/retrieve
+// GetFilesEndpoint is the OpenAI API endpoint to get files https://platform.openai.com/docs/api-reference/files/retrieve
+// @Summary Returns information about a specific file.
+// @Success 200 {object} File "Response"
+// @Router /v1/files/{file_id} [get]
 func GetFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		file, err := getFileFromRequest(c)
@@ -135,13 +138,17 @@ func GetFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Applicat
 	}
 }
 
-// DeleteFilesEndpoint https://platform.openai.com/docs/api-reference/files/delete
+type DeleteStatus struct {
+	Id      string
+	Object  string
+	Deleted bool
+}
+
+// DeleteFilesEndpoint is the OpenAI API endpoint to delete files https://platform.openai.com/docs/api-reference/files/delete
+// @Summary Delete a file.
+// @Success 200 {object} DeleteStatus "Response"
+// @Router /v1/files/{file_id} [delete]
 func DeleteFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
-	type DeleteStatus struct {
-		Id      string
-		Object  string
-		Deleted bool
-	}
 
 	return func(c *fiber.Ctx) error {
 		file, err := getFileFromRequest(c)
@@ -174,7 +181,11 @@ func DeleteFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Appli
 	}
 }
 
-// GetFilesContentsEndpoint https://platform.openai.com/docs/api-reference/files/retrieve-contents
+// GetFilesContentsEndpoint is the OpenAI API endpoint to get files content https://platform.openai.com/docs/api-reference/files/retrieve-contents
+// @Summary Returns information about a specific file.
+// @Success	200		{string}	binary				"file"
+// @Router /v1/files/{file_id}/content [get]
+// GetFilesContentsEndpoint
 func GetFilesContentsEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		file, err := getFileFromRequest(c)

core/http/endpoints/openai/files_test.go

@@ -13,10 +13,10 @@ import (
 
 	"github.com/rs/zerolog/log"
 
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/config"
 
-	utils2 "github.com/go-skynet/LocalAI/pkg/utils"
 	"github.com/gofiber/fiber/v2"
+	utils2 "github.com/mudler/LocalAI/pkg/utils"
 	"github.com/stretchr/testify/assert"
 
 	"testing"

core/http/endpoints/openai/image.go

@@ -13,14 +13,14 @@ import (
 	"strings"
 	"time"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	"github.com/go-skynet/LocalAI/core/schema"
 	"github.com/google/uuid"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
 
-	"github.com/go-skynet/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/backend"
 
-	model "github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
+	model "github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
 

core/http/endpoints/openai/inference.go

@@ -1,11 +1,11 @@
 package openai
 
 import (
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
 
-	"github.com/go-skynet/LocalAI/core/schema"
-	model "github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/schema"
+	model "github.com/mudler/LocalAI/pkg/model"
 )
 
 func ComputeChoices(

core/http/endpoints/openai/list.go

@@ -1,11 +1,15 @@
 package openai
 
 import (
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/core/services"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/services"
 )
 
+// ListModelsEndpoint is the OpenAI Models API endpoint https://platform.openai.com/docs/api-reference/models
+// @Summary List and describe the various models available in the API.
+// @Success 200 {object} schema.ModelsDataResponse "Response"
+// @Router /v1/models [get]
 func ListModelsEndpoint(lms *services.ListModelsService) func(ctx *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		// If blank, no filter is applied.
@@ -18,10 +22,7 @@ func ListModelsEndpoint(lms *services.ListModelsService) func(ctx *fiber.Ctx) er
 		if err != nil {
 			return err
 		}
-		return c.JSON(struct {
-			Object string               `json:"object"`
-			Data   []schema.OpenAIModel `json:"data"`
-		}{
+		return c.JSON(schema.ModelsDataResponse{
 			Object: "list",
 			Data:   dataModels,
 		})

core/http/endpoints/openai/request.go

@@ -2,19 +2,16 @@ package openai
 
 import (
 	"context"
-	"encoding/base64"
 	"encoding/json"
 	"fmt"
-	"io"
-	"net/http"
-	"strings"
 
-	"github.com/go-skynet/LocalAI/core/config"
-	fiberContext "github.com/go-skynet/LocalAI/core/http/ctx"
-	"github.com/go-skynet/LocalAI/core/schema"
-	"github.com/go-skynet/LocalAI/pkg/functions"
-	model "github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/config"
+	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/functions"
+	"github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/utils"
 	"github.com/rs/zerolog/log"
 )
 
@@ -39,41 +36,6 @@ func readRequest(c *fiber.Ctx, ml *model.ModelLoader, o *config.ApplicationConfi
 	return modelFile, input, err
 }
 
-// this function check if the string is an URL, if it's an URL downloads the image in memory
-// encodes it in base64 and returns the base64 string
-func getBase64Image(s string) (string, error) {
-	if strings.HasPrefix(s, "http") {
-		// download the image
-		resp, err := http.Get(s)
-		if err != nil {
-			return "", err
-		}
-		defer resp.Body.Close()
-
-		// read the image data into memory
-		data, err := io.ReadAll(resp.Body)
-		if err != nil {
-			return "", err
-		}
-
-		// encode the image data in base64
-		encoded := base64.StdEncoding.EncodeToString(data)
-
-		// return the base64 string
-		return encoded, nil
-	}
-
-	// if the string instead is prefixed with "data:image/...;base64,", drop it
-	dropPrefix := []string{"data:image/jpeg;base64,", "data:image/png;base64,"}
-	for _, prefix := range dropPrefix {
-		if strings.HasPrefix(s, prefix) {
-			return strings.ReplaceAll(s, prefix, ""), nil
-		}
-	}
-
-	return "", fmt.Errorf("not valid string")
-}
-
 func updateRequestConfig(config *config.BackendConfig, input *schema.OpenAIRequest) {
 	if input.Echo {
 		config.Echo = input.Echo
@@ -187,7 +149,7 @@ func updateRequestConfig(config *config.BackendConfig, input *schema.OpenAIReque
 					input.Messages[i].StringContent = pp.Text
 				} else if pp.Type == "image_url" {
 					// Detect if pp.ImageURL is an URL, if it is download the image and encode it in base64:
-					base64, err := getBase64Image(pp.ImageURL.URL)
+					base64, err := utils.GetImageURLAsBase64(pp.ImageURL.URL)
 					if err == nil {
 						input.Messages[i].StringImages = append(input.Messages[i].StringImages, base64) // TODO: make sure that we only return base64 stuff
 						// set a placeholder for each image
@@ -295,5 +257,9 @@ func mergeRequestWithConfig(modelFile string, input *schema.OpenAIRequest, cm *c
 	// Set the parameters for the language model prediction
 	updateRequestConfig(cfg, input)
 
+	if !cfg.Validate() {
+		return nil, nil, fmt.Errorf("failed to validate config")
+	}
+
 	return cfg, input, err
 }

core/http/endpoints/openai/transcription.go

@@ -8,9 +8,9 @@ import (
 	"path"
 	"path/filepath"
 
-	"github.com/go-skynet/LocalAI/core/backend"
-	"github.com/go-skynet/LocalAI/core/config"
-	model "github.com/go-skynet/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/core/backend"
+	"github.com/mudler/LocalAI/core/config"
+	model "github.com/mudler/LocalAI/pkg/model"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/rs/zerolog/log"
@@ -32,7 +32,7 @@ func TranscriptEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, a
 
 		config, input, err := mergeRequestWithConfig(m, input, cl, ml, appConfig.Debug, appConfig.Threads, appConfig.ContextSize, appConfig.F16)
 		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+			return fmt.Errorf("failed reading parameters from request: %w", err)
 		}
 		// retrieve the file data from the request
 		file, err := c.FormFile("file")
@@ -65,7 +65,7 @@ func TranscriptEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, a
 
 		log.Debug().Msgf("Audio file copied to: %+v", dst)
 
-		tr, err := backend.ModelTranscription(dst, input.Language, ml, *config, appConfig)
+		tr, err := backend.ModelTranscription(dst, input.Language, input.Translate, ml, *config, appConfig)
 		if err != nil {
 			return err
 		}

core/http/render.go

@@ -7,10 +7,10 @@ import (
 	"net/http"
 
 	"github.com/Masterminds/sprig/v3"
-	"github.com/go-skynet/LocalAI/core/schema"
 	"github.com/gofiber/fiber/v2"
 	fiberhtml "github.com/gofiber/template/html/v2"
 	"github.com/microcosm-cc/bluemonday"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/russross/blackfriday"
 )
 
@@ -21,14 +21,13 @@ func notFoundHandler(c *fiber.Ctx) error {
 	// Check if the request accepts JSON
 	if string(c.Context().Request.Header.ContentType()) == "application/json" || len(c.Accepts("html")) == 0 {
 		// The client expects a JSON response
-		c.Status(fiber.StatusNotFound).JSON(schema.ErrorResponse{
+		return c.Status(fiber.StatusNotFound).JSON(schema.ErrorResponse{
 			Error: &schema.APIError{Message: "Resource not found", Code: fiber.StatusNotFound},
 		})
 	} else {
 		// The client expects an HTML response
-		c.Status(fiber.StatusNotFound).Render("views/404", fiber.Map{})
+		return c.Status(fiber.StatusNotFound).Render("views/404", fiber.Map{})
 	}
-	return nil
 }
 
 func renderEngine() *fiberhtml.Engine {