Merge branch 'master' into ci/static-check

2024-12-18 20:27:57 +00:00 · 2024-07-18 19:44:59 +02:00 · 2024-07-18 19:44:59 +02:00 · c28e8ca697
commit c28e8ca697
parent ecaaff8f03 031627584b
119 changed files with 4374 additions and 1164 deletions
--- a/.github/dependabot.yml
+++ b/.github/dependabot.yml
@ -1,6 +1,10 @@
 # https://docs.github.com/en/code-security/dependabot/dependabot-version-updates/configuration-options-for-the-dependabot.yml-file
 version: 2
 updates:
+  - package-ecosystem: "gitsubmodule"
+    directory: "/"
+    schedule:
+      interval: "weekly"
  - package-ecosystem: "gomod"
    directory: "/"
    schedule:
@ -23,3 +27,111 @@ updates:
    schedule:
      # Check for updates to GitHub Actions every weekday
      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/autogptq"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/bark"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/common/template"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/coqui"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/diffusers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/exllama"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/exllama2"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/mamba"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/openvoice"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/parler-tts"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/petals"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/rerankers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/sentencetransformers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/transformers"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/transformers-musicgen"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/vall-e-x"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/backend/python/vllm"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/chainlit"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/functions"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/langchain/langchainpy-localai-example"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/langchain-chroma"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "pip"
+    directory: "/examples/streamlit-bot"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/k8sgpt"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/kubernetes"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/langchain"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "gomod"
+    directory: "/examples/semantic-todo"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "docker"
+    directory: "/examples/telegram-bot"
+    schedule:
+      interval: "weekly"
--- a/.github/workflows/bump_deps.yaml
+++ b/.github/workflows/bump_deps.yaml
@ -27,9 +27,6 @@ jobs:
          - repository: "go-skynet/bloomz.cpp"
            variable: "BLOOMZ_VERSION"
            branch: "main"
-          - repository: "nomic-ai/gpt4all"
-            variable: "GPT4ALL_VERSION"
-            branch: "main"
          - repository: "mudler/go-ggllm.cpp"
            variable: "GOGGLLM_VERSION"
            branch: "master"
@ -51,7 +48,7 @@ jobs:
          token: ${{ secrets.UPDATE_BOT_TOKEN }}
          push-to-fork: ci-forks/LocalAI
          commit-message: ':arrow_up: Update ${{ matrix.repository }}'
-          title: ':arrow_up: Update ${{ matrix.repository }}'
+          title: 'chore: :arrow_up: Update ${{ matrix.repository }}'
          branch: "update/${{ matrix.variable }}"
          body: Bump of ${{ matrix.repository }} version
          signoff: true
--- a/.github/workflows/bump_docs.yaml
+++ b/.github/workflows/bump_docs.yaml
@ -22,7 +22,7 @@ jobs:
          token: ${{ secrets.UPDATE_BOT_TOKEN }}
          push-to-fork: ci-forks/LocalAI
          commit-message: ':arrow_up: Update docs version ${{ matrix.repository }}'
-          title: ':arrow_up: Update docs version ${{ matrix.repository }}'
+          title: 'docs: :arrow_up: update docs version ${{ matrix.repository }}'
          branch: "update/docs"
          body: Bump of ${{ matrix.repository }} version inside docs
          signoff: true
--- a/.github/workflows/checksum_checker.yaml
+++ b/.github/workflows/checksum_checker.yaml
@ -25,7 +25,7 @@ jobs:
      - name: 'Setup yq'
        uses: dcarbone/install-yq-action@v1.1.1
        with:
-          version: 'v4.43.1'
+          version: 'v4.44.2'
          download-compressed: true
          force: true

--- a/.github/workflows/comment-pr.yaml
+++ b/.github/workflows/comment-pr.yaml
@ -0,0 +1,81 @@
+name: Comment PRs
+on:
+  pull_request_target:
+
+jobs:
+  comment-pr:
+    env:
+        MODEL_NAME: hermes-2-theta-llama-3-8b
+    runs-on: ubuntu-latest
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v3
+      with:
+        ref: "${{ github.event.pull_request.merge_commit_sha }}"
+    - uses: mudler/localai-github-action@v1
+      with:
+        model: 'hermes-2-theta-llama-3-8b' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
+      # Check the PR diff using the current branch and the base branch of the PR
+    - uses: GrantBirki/git-diff-action@v2.7.0
+      id: git-diff-action
+      with:
+            json_diff_file_output: diff.json
+            raw_diff_file_output: diff.txt
+            file_output_only: "true"
+    - name: Show diff
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      run: |
+            cat $DIFF
+    - name: Summarize
+      env:
+        DIFF: ${{ steps.git-diff-action.outputs.raw-diff-path }}
+      id: summarize
+      run: |
+            input="$(cat $DIFF)"
+
+            # Define the LocalAI API endpoint
+            API_URL="http://localhost:8080/chat/completions"
+
+            # Create a JSON payload using jq to handle special characters
+            json_payload=$(jq -n --arg input "$input" '{
+            model: "'$MODEL_NAME'",
+            messages: [
+                {
+                role: "system",
+                content: "You are LocalAI-bot in Github that helps understanding PRs and assess complexity. Explain what has changed in this PR diff and why"
+                },
+                {
+                role: "user",
+                content: $input
+                }
+            ]
+            }')
+
+            # Send the request to LocalAI
+            response=$(curl -s -X POST $API_URL \
+            -H "Content-Type: application/json" \
+            -d "$json_payload")
+
+            # Extract the summary from the response
+            summary="$(echo $response | jq -r '.choices[0].message.content')"
+
+            # Print the summary
+            #  -H "Authorization: Bearer $API_KEY" \
+            echo "Summary:"
+            echo "$summary"
+            echo "payload sent"
+            echo "$json_payload"
+            {
+                echo 'message<<EOF'
+                echo "$summary"
+                echo EOF
+              } >> "$GITHUB_OUTPUT"
+            docker logs --tail 10 local-ai
+    - uses: mshick/add-pr-comment@v2
+      if: always()
+      with:
+          repo-token: ${{ secrets.UPDATE_BOT_TOKEN }}
+          message: ${{ steps.summarize.outputs.message }}
+          message-failure: |
+            Uh oh! Could not analyze this PR, maybe it's too big?
--- a/.github/workflows/dependabot_auto.yml
+++ b/.github/workflows/dependabot_auto.yml
@ -14,7 +14,7 @@ jobs:
    steps:
      - name: Dependabot metadata
        id: metadata
-        uses: dependabot/fetch-metadata@v2.1.0
+        uses: dependabot/fetch-metadata@v2.2.0
        with:
          github-token: "${{ secrets.GITHUB_TOKEN }}"
          skip-commit-verification: true
--- a/.github/workflows/generate_grpc_cache.yaml
+++ b/.github/workflows/generate_grpc_cache.yaml
@ -75,7 +75,7 @@ jobs:
        uses: actions/checkout@v4

      - name: Cache GRPC
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
        with:
          builder: ${{ steps.buildx.outputs.name }}
          # The build-args MUST be an EXACT match between the image cache and other workflow steps that want to use that cache.
@ -84,7 +84,7 @@ jobs:
          build-args: |
            GRPC_BASE_IMAGE=${{ matrix.grpc-base-image }}
            GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
-            GRPC_VERSION=v1.64.0
+            GRPC_VERSION=v1.65.0
          context: .
          file: ./Dockerfile
          cache-to: type=gha,ignore-error=true
--- a/.github/workflows/generate_intel_image.yaml
+++ b/.github/workflows/generate_intel_image.yaml
@ -15,7 +15,7 @@ jobs:
    strategy:
      matrix:
        include:
-          - base-image: intel/oneapi-basekit:2024.1.0-devel-ubuntu22.04
+          - base-image: intel/oneapi-basekit:2024.2.0-devel-ubuntu22.04
            runs-on: 'ubuntu-latest'
            platforms: 'linux/amd64'
    runs-on: ${{matrix.runs-on}}
@ -46,7 +46,7 @@ jobs:
        uses: actions/checkout@v4

      - name: Cache Intel images
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
        with:
          builder: ${{ steps.buildx.outputs.name }}
          build-args: |
--- a/.github/workflows/image_build.yml
+++ b/.github/workflows/image_build.yml
@ -215,7 +215,7 @@ jobs:
          password: ${{ secrets.quayPassword }}

      - name: Build and push
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
        if: github.event_name != 'pull_request'
        with:
          builder: ${{ steps.buildx.outputs.name }}
@ -232,7 +232,7 @@ jobs:
            BASE_IMAGE=${{ inputs.base-image }}
            GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
            GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
-            GRPC_VERSION=v1.64.0
+            GRPC_VERSION=v1.65.0
            MAKEFLAGS=${{ inputs.makeflags }}
          context: .
          file: ./Dockerfile
@ -243,7 +243,7 @@ jobs:
          labels: ${{ steps.meta.outputs.labels }}
 ### Start testing image
      - name: Build and push
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
        if: github.event_name == 'pull_request'
        with:
          builder: ${{ steps.buildx.outputs.name }}
@ -260,7 +260,7 @@ jobs:
            BASE_IMAGE=${{ inputs.base-image }}
            GRPC_BASE_IMAGE=${{ inputs.grpc-base-image || inputs.base-image }}
            GRPC_MAKEFLAGS=--jobs=4 --output-sync=target
-            GRPC_VERSION=v1.64.0
+            GRPC_VERSION=v1.65.0
            MAKEFLAGS=${{ inputs.makeflags }}
          context: .
          file: ./Dockerfile
@ -276,7 +276,7 @@ jobs:
 ## End testing image
      - name: Build and push AIO image
        if: inputs.aio != ''
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
        with:
          builder: ${{ steps.buildx.outputs.name }}
          build-args: |
@ -291,7 +291,7 @@ jobs:

      - name: Build and push AIO image (dockerhub)
        if: inputs.aio != ''
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
        with:
          builder: ${{ steps.buildx.outputs.name }}
          build-args: |
--- a/.github/workflows/notify-models.yaml
+++ b/.github/workflows/notify-models.yaml
@ -14,11 +14,9 @@ jobs:
    - 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
+    - uses: mudler/localai-github-action@v1
+      with:
+        model: 'hermes-2-theta-llama-3-8b' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
        # Check the PR diff using the current branch and the base branch of the PR
    - uses: GrantBirki/git-diff-action@v2.7.0
      id: git-diff-action
--- a/.github/workflows/notify-releases.yaml
+++ b/.github/workflows/notify-releases.yaml
@ -12,11 +12,9 @@ jobs:
        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
+    - uses: mudler/localai-github-action@v1
+      with:
+        model: 'hermes-2-theta-llama-3-8b' # Any from models.localai.io, or from huggingface.com with: "huggingface://<repository>/file"
    - name: Summarize
      id: summarize
      run: |
--- a/.github/workflows/prlint.yaml
+++ b/.github/workflows/prlint.yaml
@ -17,12 +17,12 @@ jobs:
      - uses: aslafy-z/conventional-pr-title-action@v3
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-  check-pr-description:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v2
-      - uses: jadrol/pr-description-checker-action@v1.0.0
-        id: description-checker
-        with:
-          repo-token: ${{ secrets.GITHUB_TOKEN }}
-          exempt-labels: no qa
+#  check-pr-description:
+#    runs-on: ubuntu-latest
+#    steps:
+#      - uses: actions/checkout@v2
+#      - uses: jadrol/pr-description-checker-action@v1.0.0
+#        id: description-checker
+#        with:
+#          repo-token: ${{ secrets.GITHUB_TOKEN }}
+#          exempt-labels: no qa
--- a/.github/workflows/release.yaml
+++ b/.github/workflows/release.yaml
@ -7,7 +7,7 @@ on:
  pull_request:

 env:
-  GRPC_VERSION: v1.64.0
+  GRPC_VERSION: v1.65.0

 permissions:
  contents: write
@ -99,8 +99,8 @@ 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
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
          export PATH=$PATH:$GOPATH/bin
          export PATH=/usr/local/cuda/bin:$PATH
          sudo rm -rf /usr/aarch64-linux-gnu/lib/libstdc++.so.6
@ -163,7 +163,7 @@ jobs:
          sudo apt-get update
          sudo apt-get install -y cuda-nvcc-${CUDA_VERSION} libcublas-dev-${CUDA_VERSION}
        env:
-          CUDA_VERSION: 12-3
+          CUDA_VERSION: 12-5
      - name: "Install Hipblas"
        env:
          ROCM_VERSION: "6.1"
@ -210,8 +210,8 @@ jobs:
      - name: Build
        id: build
        run: |
-          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
-          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
          export PATH=$PATH:$GOPATH/bin
          export PATH=/usr/local/cuda/bin:$PATH
          export PATH=/opt/rocm/bin:$PATH
@ -251,8 +251,8 @@ jobs:
        run: |
          sudo apt-get update
          sudo apt-get install -y --no-install-recommends libopencv-dev protobuf-compiler ccache
-          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
-          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
      - name: Build stablediffusion
        run: |
          export PATH=$PATH:$GOPATH/bin
@ -327,8 +327,8 @@ jobs:
      - 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
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
      - name: Build
        id: build
        run: |
--- a/.github/workflows/test-extra.yml
+++ b/.github/workflows/test-extra.yml
@ -29,7 +29,7 @@ jobs:
          curl -LsSf https://astral.sh/uv/install.sh | sh
          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
          sudo apt-get install -y libopencv-dev
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1

      - name: Test transformers
        run: |
@ -51,7 +51,7 @@ jobs:
          curl -LsSf https://astral.sh/uv/install.sh | sh
          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
          sudo apt-get install -y libopencv-dev
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1

      - name: Test sentencetransformers
        run: |
@ -74,7 +74,7 @@ jobs:
          curl -LsSf https://astral.sh/uv/install.sh | sh
          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
          sudo apt-get install -y libopencv-dev
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1

      - name: Test rerankers
        run: |
@ -96,7 +96,7 @@ jobs:
          sudo apt-get install -y libopencv-dev
          # Install UV
          curl -LsSf https://astral.sh/uv/install.sh | sh
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
      - name: Test diffusers
        run: |
          make --jobs=5 --output-sync=target -C backend/python/diffusers
@ -117,7 +117,7 @@ jobs:
          curl -LsSf https://astral.sh/uv/install.sh | sh
          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
          sudo apt-get install -y libopencv-dev
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1

      - name: Test parler-tts
        run: |
@ -139,7 +139,7 @@ jobs:
          curl -LsSf https://astral.sh/uv/install.sh | sh
          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
          sudo apt-get install -y libopencv-dev
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1

      - name: Test openvoice
        run: |
@ -161,7 +161,7 @@ jobs:
          curl -LsSf https://astral.sh/uv/install.sh | sh
          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
          sudo apt-get install -y libopencv-dev
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1

      - name: Test transformers-musicgen
        run: |
@ -185,7 +185,7 @@ jobs:
  #         curl -LsSf https://astral.sh/uv/install.sh | sh
  #         sudo apt-get install -y ca-certificates cmake curl patch python3-pip
  #         sudo apt-get install -y libopencv-dev
-  #         pip install --user grpcio-tools==1.64.0
+  #         pip install --user --no-cache-dir grpcio-tools==1.64.1

  #     - name: Test petals
  #       run: |
@ -249,7 +249,7 @@ jobs:
  #         curl -LsSf https://astral.sh/uv/install.sh | sh
  #         sudo apt-get install -y ca-certificates cmake curl patch python3-pip
  #         sudo apt-get install -y libopencv-dev
-  #         pip install --user grpcio-tools==1.64.0
+  #         pip install --user --no-cache-dir grpcio-tools==1.64.1

  #     - name: Test bark
  #       run: |
@ -274,7 +274,7 @@ jobs:
  #         curl -LsSf https://astral.sh/uv/install.sh | sh
  #         sudo apt-get install -y ca-certificates cmake curl patch python3-pip
  #         sudo apt-get install -y libopencv-dev
-  #         pip install --user grpcio-tools==1.64.0
+  #         pip install --user --no-cache-dir grpcio-tools==1.64.1
  #     - name: Test vllm
  #       run: |
  #          make --jobs=5 --output-sync=target -C backend/python/vllm
@ -294,7 +294,7 @@ jobs:
          curl -LsSf https://astral.sh/uv/install.sh | sh
          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
          sudo apt-get install -y libopencv-dev
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
      - name: Test vall-e-x
        run: |
           make --jobs=5 --output-sync=target -C backend/python/vall-e-x
@ -314,7 +314,7 @@ jobs:
          sudo apt-get install -y ca-certificates cmake curl patch espeak espeak-ng python3-pip
          # Install UV
          curl -LsSf https://astral.sh/uv/install.sh | sh
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
      - name: Test coqui
        run: |
          make --jobs=5 --output-sync=target -C backend/python/coqui
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@ -10,7 +10,7 @@ on:
      - '*'

 env:
-  GRPC_VERSION: v1.64.0
+  GRPC_VERSION: v1.65.0

 concurrency:
  group: ci-tests-${{ github.head_ref || github.ref }}-${{ github.repository }}
@ -94,8 +94,8 @@ jobs:
          sudo apt-get install -y cuda-nvcc-${CUDA_VERSION} libcublas-dev-${CUDA_VERSION}
          export CUDACXX=/usr/local/cuda/bin/nvcc

-          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
-          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af

          # The python3-grpc-tools package in 22.04 is too old
          pip install --user grpcio-tools
@ -110,7 +110,7 @@ jobs:
          # Pre-build stable diffusion before we install a newer version of abseil (not compatible with stablediffusion-ncn)
          PATH="$PATH:/root/go/bin" GO_TAGS="stablediffusion tts" GRPC_BACKENDS=backend-assets/grpc/stablediffusion make build
        env:
-          CUDA_VERSION: 12-3
+          CUDA_VERSION: 12-4
      - name: Cache grpc
        id: cache-grpc
        uses: actions/cache@v4
@ -215,7 +215,7 @@ jobs:
      - name: Dependencies
        run: |
          brew install protobuf grpc make protoc-gen-go protoc-gen-go-grpc
-          pip install --user grpcio-tools==1.64.0
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
      - name: Test
        run: |
          export C_INCLUDE_PATH=/usr/local/include
--- a/.github/workflows/update_swagger.yaml
+++ b/.github/workflows/update_swagger.yaml
@ -13,11 +13,17 @@ jobs:
      - uses: actions/setup-go@v5
        with:
          go-version: 'stable'
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install protobuf-compiler
      - run: |
          go install github.com/swaggo/swag/cmd/swag@latest
+          go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
+          go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2
      - name: Bump swagger 🔧
        run: |
-          make swagger
+          make protogen-go swagger
      - name: Create Pull Request
        uses: peter-evans/create-pull-request@v6
        with:
--- a/23
+++ b/23
@ -8,7 +8,7 @@ FROM ${BASE_IMAGE} AS requirements-core

 USER root

-ARG GO_VERSION=1.22.4
+ARG GO_VERSION=1.22.5
 ARG TARGETARCH
 ARG TARGETVARIANT

@ -146,25 +146,6 @@ RUN <<EOT bash
    fi
 EOT

-RUN if [ "${BUILD_TYPE}" = "cublas" ]; then \
-        apt-get update && \
-        apt-get install -y  --no-install-recommends \
-            software-properties-common pciutils && \
-        curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb && \
-        dpkg -i cuda-keyring_1.1-1_all.deb && \
-        rm -f cuda-keyring_1.1-1_all.deb && \
-        apt-get update && \
-        apt-get install -y --no-install-recommends \
-            cuda-nvcc-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcufft-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcurand-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcublas-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcusparse-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
-            libcusolver-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} && \
-        apt-get clean && \
-        rm -rf /var/lib/apt/lists/* \
-    ; fi
-
 # If we are building with clblas support, we need the libraries for the builds
 RUN if [ "${BUILD_TYPE}" = "clblas" ]; then \
        apt-get update && \
@ -206,7 +187,7 @@ FROM ${GRPC_BASE_IMAGE} AS grpc

 # This is a bit of a hack, but it's required in order to be able to effectively cache this layer in CI
 ARG GRPC_MAKEFLAGS="-j4 -Otarget"
-ARG GRPC_VERSION=v1.64.2
+ARG GRPC_VERSION=v1.65.0

 ENV MAKEFLAGS=${GRPC_MAKEFLAGS}

--- a/46
+++ b/46
@ -8,7 +8,7 @@ DETECT_LIBS?=true
 # llama.cpp versions
 GOLLAMA_REPO?=https://github.com/go-skynet/go-llama.cpp
 GOLLAMA_VERSION?=2b57a8ae43e4699d3dc5d1496a1ccd42922993be
-CPPLLAMA_VERSION?=368645698ab648e390dcd7c00a2bf60efa654f57
+CPPLLAMA_VERSION?=b3283448ce9a5098226afe1d8648ccc578511fe4

 # gpt4all version
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
@ -20,7 +20,7 @@ RWKV_VERSION?=661e7ae26d442f5cfebd2a0881b44e8c55949ec6

 # whisper.cpp version
 WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp
-WHISPER_CPP_VERSION?=d207c6882247984689091ae9d780d2e51eab1df7
+WHISPER_CPP_VERSION?=f68298ce06ca3edd6e6f3f21c3d0bb5f073942c3

 # bert.cpp version
 BERT_REPO?=https://github.com/go-skynet/go-bert.cpp
@ -214,7 +214,7 @@ sources/go-bert.cpp:
 	git remote add origin $(BERT_REPO) && \
 	git fetch origin && \
 	git checkout $(BERT_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/go-bert.cpp/libgobert.a: sources/go-bert.cpp
 	$(MAKE) -C sources/go-bert.cpp libgobert.a
@ -227,7 +227,7 @@ sources/go-llama.cpp:
 	git remote add origin $(GOLLAMA_REPO) && \
 	git fetch origin && \
 	git checkout $(GOLLAMA_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/go-llama.cpp/libbinding.a: sources/go-llama.cpp
 	$(MAKE) -C sources/go-llama.cpp BUILD_TYPE=$(STABLE_BUILD_TYPE) libbinding.a
@ -240,7 +240,7 @@ sources/go-piper:
 	git remote add origin $(PIPER_REPO) && \
 	git fetch origin && \
 	git checkout $(PIPER_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/go-piper/libpiper_binding.a: sources/go-piper
 	$(MAKE) -C sources/go-piper libpiper_binding.a example/main piper.o
@ -253,7 +253,7 @@ sources/gpt4all:
 	git remote add origin $(GPT4ALL_REPO) && \
 	git fetch origin && \
 	git checkout $(GPT4ALL_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/gpt4all/gpt4all-bindings/golang/libgpt4all.a: sources/gpt4all
 	$(MAKE) -C sources/gpt4all/gpt4all-bindings/golang/ libgpt4all.a
@ -266,7 +266,7 @@ sources/go-rwkv.cpp:
 	git remote add origin $(RWKV_REPO) && \
 	git fetch origin && \
 	git checkout $(RWKV_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/go-rwkv.cpp/librwkv.a: sources/go-rwkv.cpp
 	cd sources/go-rwkv.cpp && cd rwkv.cpp &&	cmake . -DRWKV_BUILD_SHARED_LIBRARY=OFF &&	cmake --build . && 	cp librwkv.a ..
@ -279,7 +279,7 @@ sources/go-stable-diffusion:
 	git remote add origin $(STABLEDIFFUSION_REPO) && \
 	git fetch origin && \
 	git checkout $(STABLEDIFFUSION_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/go-stable-diffusion/libstablediffusion.a: sources/go-stable-diffusion
 	CPATH="$(CPATH):/usr/include/opencv4" $(MAKE) -C sources/go-stable-diffusion libstablediffusion.a
@ -292,7 +292,7 @@ sources/go-tiny-dream:
 	git remote add origin $(TINYDREAM_REPO) && \
 	git fetch origin && \
 	git checkout $(TINYDREAM_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/go-tiny-dream/libtinydream.a: sources/go-tiny-dream
 	$(MAKE) -C sources/go-tiny-dream libtinydream.a
@ -305,12 +305,12 @@ sources/whisper.cpp:
 	git remote add origin $(WHISPER_REPO) && \
 	git fetch origin && \
 	git checkout $(WHISPER_CPP_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 sources/whisper.cpp/libwhisper.a: sources/whisper.cpp
 	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
+get-sources: sources/go-llama.cpp sources/gpt4all sources/go-piper sources/go-rwkv.cpp sources/whisper.cpp sources/go-bert.cpp sources/go-stable-diffusion sources/go-tiny-dream backend/cpp/llama/llama.cpp

 replace:
 	$(GOCMD) mod edit -replace github.com/donomii/go-rwkv.cpp=$(CURDIR)/sources/go-rwkv.cpp
@ -384,7 +384,7 @@ endif
 	CGO_LDFLAGS="$(CGO_LDFLAGS)" $(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o $(BINARY_NAME) ./

 build-minimal:
-	BUILD_GRPC_FOR_BACKEND_LLAMA=true GRPC_BACKENDS="backend-assets/grpc/llama-cpp-avx2" GO_TAGS=none $(MAKE) build
+	BUILD_GRPC_FOR_BACKEND_LLAMA=true GRPC_BACKENDS="backend-assets/grpc/llama-cpp-avx2" GO_TAGS=p2p $(MAKE) build

 build-api:
 	BUILD_GRPC_FOR_BACKEND_LLAMA=true BUILD_API_ONLY=true GO_TAGS=none $(MAKE) build
@ -767,28 +767,28 @@ else
 endif

 # This target is for manually building a variant with-auto detected flags
-backend-assets/grpc/llama-cpp: backend-assets/grpc
+backend-assets/grpc/llama-cpp: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-cpp
 	$(MAKE) -C backend/cpp/llama-cpp purge
 	$(info ${GREEN}I llama-cpp build info:avx2${RESET})
 	$(MAKE) VARIANT="llama-cpp" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-cpp/grpc-server backend-assets/grpc/llama-cpp

-backend-assets/grpc/llama-cpp-avx2: backend-assets/grpc
+backend-assets/grpc/llama-cpp-avx2: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-avx2
 	$(MAKE) -C backend/cpp/llama-avx2 purge
 	$(info ${GREEN}I llama-cpp build info:avx2${RESET})
 	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=on -DGGML_AVX512=off -DGGML_FMA=on -DGGML_F16C=on" $(MAKE) VARIANT="llama-avx2" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-avx2/grpc-server backend-assets/grpc/llama-cpp-avx2

-backend-assets/grpc/llama-cpp-avx: backend-assets/grpc
+backend-assets/grpc/llama-cpp-avx: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-avx
 	$(MAKE) -C backend/cpp/llama-avx purge
 	$(info ${GREEN}I llama-cpp build info:avx${RESET})
 	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off" $(MAKE) VARIANT="llama-avx" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-avx/grpc-server backend-assets/grpc/llama-cpp-avx

-backend-assets/grpc/llama-cpp-fallback: backend-assets/grpc
+backend-assets/grpc/llama-cpp-fallback: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-fallback
 	$(MAKE) -C backend/cpp/llama-fallback purge
 	$(info ${GREEN}I llama-cpp build info:fallback${RESET})
@ -799,35 +799,35 @@ ifeq ($(BUILD_TYPE),metal)
 	cp backend/cpp/llama-fallback/llama.cpp/build/bin/default.metallib backend-assets/grpc/
 endif

-backend-assets/grpc/llama-cpp-cuda: backend-assets/grpc
+backend-assets/grpc/llama-cpp-cuda: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-cuda
 	$(MAKE) -C backend/cpp/llama-cuda purge
 	$(info ${GREEN}I llama-cpp build info:cuda${RESET})
 	CMAKE_ARGS="$(CMAKE_ARGS) -DGGML_AVX=on -DGGML_AVX2=off -DGGML_AVX512=off -DGGML_FMA=off -DGGML_F16C=off -DGGML_CUDA=ON" $(MAKE) VARIANT="llama-cuda" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-cuda/grpc-server backend-assets/grpc/llama-cpp-cuda

-backend-assets/grpc/llama-cpp-hipblas: backend-assets/grpc
+backend-assets/grpc/llama-cpp-hipblas: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-hipblas
 	$(MAKE) -C backend/cpp/llama-hipblas purge
 	$(info ${GREEN}I llama-cpp build info:hipblas${RESET})
 	BUILD_TYPE="hipblas" $(MAKE) VARIANT="llama-hipblas" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-hipblas/grpc-server backend-assets/grpc/llama-cpp-hipblas

-backend-assets/grpc/llama-cpp-sycl_f16: backend-assets/grpc
+backend-assets/grpc/llama-cpp-sycl_f16: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-sycl_f16
 	$(MAKE) -C backend/cpp/llama-sycl_f16 purge
 	$(info ${GREEN}I llama-cpp build info:sycl_f16${RESET})
 	BUILD_TYPE="sycl_f16" $(MAKE) VARIANT="llama-sycl_f16" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-sycl_f16/grpc-server backend-assets/grpc/llama-cpp-sycl_f16

-backend-assets/grpc/llama-cpp-sycl_f32: backend-assets/grpc
+backend-assets/grpc/llama-cpp-sycl_f32: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-sycl_f32
 	$(MAKE) -C backend/cpp/llama-sycl_f32 purge
 	$(info ${GREEN}I llama-cpp build info:sycl_f32${RESET})
 	BUILD_TYPE="sycl_f32" $(MAKE) VARIANT="llama-sycl_f32" build-llama-cpp-grpc-server
 	cp -rfv backend/cpp/llama-sycl_f32/grpc-server backend-assets/grpc/llama-cpp-sycl_f32

-backend-assets/grpc/llama-cpp-grpc: backend-assets/grpc
+backend-assets/grpc/llama-cpp-grpc: backend-assets/grpc backend/cpp/llama/llama.cpp
 	cp -rf backend/cpp/llama backend/cpp/llama-grpc
 	$(MAKE) -C backend/cpp/llama-grpc purge
 	$(info ${GREEN}I llama-cpp build info:grpc${RESET})
@ -905,7 +905,7 @@ docker-aio-all:

 docker-image-intel:
 	docker build \
-		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.1.0-devel-ubuntu22.04 \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.2.0-devel-ubuntu22.04 \
 		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
 		--build-arg GO_TAGS="none" \
 		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
@ -913,7 +913,7 @@ docker-image-intel:

 docker-image-intel-xpu:
 	docker build \
-		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.1.0-devel-ubuntu22.04 \
+		--build-arg BASE_IMAGE=intel/oneapi-basekit:2024.2.0-devel-ubuntu22.04 \
 		--build-arg IMAGE_TYPE=$(IMAGE_TYPE) \
 		--build-arg GO_TAGS="none" \
 		--build-arg MAKEFLAGS="$(DOCKER_MAKEFLAGS)" \
--- a/README.md
+++ b/README.md
@ -72,14 +72,15 @@ 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
- 🔥🔥 Distributed inferencing: https://github.com/mudler/LocalAI/pull/2324
- Chat, TTS, and Image generation in the WebUI: https://github.com/mudler/LocalAI/pull/2222
- Reranker API: https://github.com/mudler/LocalAI/pull/2121
+- July 2024: 🔥🔥 🆕 P2P Dashboard, LocalAI Federated mode and AI Swarms: https://github.com/mudler/LocalAI/pull/2723
+- June 2024: 🆕 You can browse now the model gallery without LocalAI! Check out https://models.localai.io
+- June 2024: Support for models from OCI registries: https://github.com/mudler/LocalAI/pull/2628
+- May 2024: 🔥🔥 Decentralized P2P llama.cpp:  https://github.com/mudler/LocalAI/pull/2343 (peer2peer llama.cpp!) 👉 Docs  https://localai.io/features/distribute/
+- May 2024: 🔥🔥 Openvoice: https://github.com/mudler/LocalAI/pull/2334
+- May 2024: 🆕 Function calls without grammars and mixed mode: https://github.com/mudler/LocalAI/pull/2328
+- May 2024: 🔥🔥 Distributed inferencing: https://github.com/mudler/LocalAI/pull/2324
+- May 2024: Chat, TTS, and Image generation in the WebUI: https://github.com/mudler/LocalAI/pull/2222
+- April 2024: Reranker API: https://github.com/mudler/LocalAI/pull/2121

 Hot topics (looking for contributors):

@ -89,6 +90,7 @@ Hot topics (looking for contributors):
 - Assistant API: https://github.com/mudler/LocalAI/issues/1273
 - Moderation endpoint: https://github.com/mudler/LocalAI/issues/999
 - Vulkan: https://github.com/mudler/LocalAI/issues/1647
+- Anthropic API: https://github.com/mudler/LocalAI/issues/1808

 If you want to help and contribute, issues up for grabs: https://github.com/mudler/LocalAI/issues?q=is%3Aissue+is%3Aopen+label%3A%22up+for+grabs%22

@ -134,6 +136,7 @@ Other:
 - Slack bot https://github.com/mudler/LocalAGI/tree/main/examples/slack
 - Shell-Pilot(Interact with LLM using LocalAI models via pure shell scripts on your Linux or MacOS system) https://github.com/reid41/shell-pilot
 - Telegram bot https://github.com/mudler/LocalAI/tree/master/examples/telegram-bot
+- Github Actions: https://github.com/marketplace/actions/start-localai
 - Examples: https://github.com/mudler/LocalAI/tree/master/examples/
  

--- a/backend/cpp/grpc/Makefile
+++ b/backend/cpp/grpc/Makefile
@ -52,7 +52,7 @@ $(GRPC_REPO):
 	git remote add origin $(GIT_REPO_LIB_GRPC)  && \
 	git fetch origin && \
 	git checkout $(TAG_LIB_GRPC) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch
 	
 $(GRPC_BUILD): $(GRPC_REPO)
 	mkdir -p $(GRPC_BUILD)
--- a/backend/cpp/llama/Makefile
+++ b/backend/cpp/llama/Makefile
@ -52,7 +52,7 @@ llama.cpp:
 	git remote add origin $(LLAMA_REPO)  && \
 	git fetch origin && \
 	git checkout -b build $(LLAMA_VERSION) && \
-	git submodule update --init --recursive --depth 1
+	git submodule update --init --recursive --depth 1 --single-branch

 llama.cpp/examples/grpc-server: llama.cpp
 	mkdir -p llama.cpp/examples/grpc-server
--- a/backend/cpp/llama/grpc-server.cpp
+++ b/backend/cpp/llama/grpc-server.cpp
@ -2108,6 +2108,7 @@ json parse_options(bool streaming, const backend::PredictOptions* predict, llama
    data["grammar"] = predict->grammar();
    data["prompt"] = predict->prompt();
    data["ignore_eos"] = predict->ignoreeos();
+    data["embeddings"] = predict->embeddings();

    // for each image in the request, add the image data
    //
@ -2385,6 +2386,31 @@ public:

        return grpc::Status::OK;
    }
+
+    /// https://github.com/ggerganov/llama.cpp/blob/aa2341298924ac89778252015efcb792f2df1e20/examples/server/server.cpp#L2969
+    grpc::Status Embedding(ServerContext* context, const backend::PredictOptions* request, backend::EmbeddingResult* embeddingResult) {
+        json data = parse_options(false, request, llama);
+        const int task_id = llama.queue_tasks.get_new_id();
+        llama.queue_results.add_waiting_task_id(task_id);
+        llama.request_completion(task_id, { {"prompt", data["embeddings"]}, { "n_predict", 0}, {"image_data", ""} }, false, true, -1);
+        // get the result
+        task_result result = llama.queue_results.recv(task_id);
+        //std::cout << "Embedding result JSON" << result.result_json.dump() << std::endl;
+        llama.queue_results.remove_waiting_task_id(task_id);
+        if (!result.error && result.stop) {
+            std::vector<float> embeddings = result.result_json.value("embedding", std::vector<float>());
+            // loop the vector and set the embeddings results
+            for (int i = 0; i < embeddings.size(); i++) {
+                embeddingResult->add_embeddings(embeddings[i]);
+            }
+        }
+        else
+        {
+            return grpc::Status::OK;
+        }
+
+        return grpc::Status::OK;
+    }
 };

 void RunServer(const std::string& server_address) {
--- a/backend/go/llm/llama/llama.go
+++ b/backend/go/llm/llama/llama.go
@ -6,9 +6,9 @@ import (
 	"fmt"
 	"path/filepath"

-	pb "github.com/go-skynet/LocalAI/pkg/grpc/proto"
 	"github.com/go-skynet/go-llama.cpp"
 	"github.com/mudler/LocalAI/pkg/grpc/base"
+	pb "github.com/mudler/LocalAI/pkg/grpc/proto"
 )

 type LLM struct {
--- a/backend/python/autogptq/requirements-intel.txt
+++ b/backend/python/autogptq/requirements-intel.txt
@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/autogptq/requirements.txt
+++ b/backend/python/autogptq/requirements.txt
@ -1,6 +1,6 @@
 accelerate
 auto-gptq==0.7.1
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 torch
 certifi
--- a/backend/python/bark/requirements-intel.txt
+++ b/backend/python/bark/requirements-intel.txt
@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/bark/requirements.txt
+++ b/backend/python/bark/requirements.txt
@ -1,6 +1,6 @@
 accelerate
 bark==0.1.5
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 certifi
 transformers
--- a/backend/python/common/template/requirements.txt
+++ b/backend/python/common/template/requirements.txt
@ -1,2 +1,2 @@
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
--- a/backend/python/coqui/requirements-intel.txt
+++ b/backend/python/coqui/requirements-intel.txt
@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/coqui/requirements.txt
+++ b/backend/python/coqui/requirements.txt
@ -1,6 +1,6 @@
 accelerate
 TTS==0.22.0
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 certifi
 transformers
--- a/backend/python/diffusers/backend.py
+++ b/backend/python/diffusers/backend.py
@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 from concurrent import futures
-
+import traceback
 import argparse
 from collections import defaultdict
 from enum import Enum
@ -17,35 +17,39 @@ import backend_pb2_grpc

 import grpc

-from diffusers import StableDiffusion3Pipeline, StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, EulerAncestralDiscreteScheduler
+from diffusers import StableDiffusion3Pipeline, StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, \
+    EulerAncestralDiscreteScheduler
 from diffusers import StableDiffusionImg2ImgPipeline, AutoPipelineForText2Image, ControlNetModel, StableVideoDiffusionPipeline
 from diffusers.pipelines.stable_diffusion import safety_checker
-from diffusers.utils import load_image,export_to_video
+from diffusers.utils import load_image, export_to_video
 from compel import Compel, ReturnedEmbeddingsType

 from transformers import CLIPTextModel
 from safetensors.torch import load_file

-
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
-COMPEL=os.environ.get("COMPEL", "0") == "1"
-XPU=os.environ.get("XPU", "0") == "1"
-CLIPSKIP=os.environ.get("CLIPSKIP", "1") == "1"
-SAFETENSORS=os.environ.get("SAFETENSORS", "1") == "1"
-CHUNK_SIZE=os.environ.get("CHUNK_SIZE", "8")
-FPS=os.environ.get("FPS", "7")
-DISABLE_CPU_OFFLOAD=os.environ.get("DISABLE_CPU_OFFLOAD", "0") == "1"
-FRAMES=os.environ.get("FRAMES", "64")
+COMPEL = os.environ.get("COMPEL", "0") == "1"
+XPU = os.environ.get("XPU", "0") == "1"
+CLIPSKIP = os.environ.get("CLIPSKIP", "1") == "1"
+SAFETENSORS = os.environ.get("SAFETENSORS", "1") == "1"
+CHUNK_SIZE = os.environ.get("CHUNK_SIZE", "8")
+FPS = os.environ.get("FPS", "7")
+DISABLE_CPU_OFFLOAD = os.environ.get("DISABLE_CPU_OFFLOAD", "0") == "1"
+FRAMES = os.environ.get("FRAMES", "64")

 if XPU:
    import intel_extension_for_pytorch as ipex
+
    print(ipex.xpu.get_device_name(0))

 # If MAX_WORKERS are specified in the environment use it, otherwise default to 1
 MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))

+
 # https://github.com/CompVis/stable-diffusion/issues/239#issuecomment-1627615287
-def sc(self, clip_input, images) : return images, [False for i in images]
+def sc(self, clip_input, images): return images, [False for i in images]
+
+
 # edit the StableDiffusionSafetyChecker class so that, when called, it just returns the images and an array of True values
 safety_checker.StableDiffusionSafetyChecker.forward = sc

@ -62,6 +66,8 @@ from diffusers.schedulers import (
    PNDMScheduler,
    UniPCMultistepScheduler,
 )
+
+
 # The scheduler list mapping was taken from here: https://github.com/neggles/animatediff-cli/blob/6f336f5f4b5e38e85d7f06f1744ef42d0a45f2a7/src/animatediff/schedulers.py#L39
 # Credits to https://github.com/neggles
 # See https://github.com/huggingface/diffusers/issues/4167 for more details on sched mapping from A1111
@ -136,10 +142,12 @@ def get_scheduler(name: str, config: dict = {}):

    return sched_class.from_config(config)

+
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
    def Health(self, request, context):
        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
+
    def LoadModel(self, request, context):
        try:
            print(f"Loading model {request.Model}...", file=sys.stderr)
@ -149,7 +157,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):

            if request.F16Memory:
                torchType = torch.float16
-                variant="fp16"
+                variant = "fp16"

            local = False
            modelFile = request.Model
@ -172,8 +180,8 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                    modelFile = request.ModelFile

            fromSingleFile = request.Model.startswith("http") or request.Model.startswith("/") or local
-            self.img2vid=False
-            self.txt2vid=False
+            self.img2vid = False
+            self.txt2vid = False
            ## img2img
            if (request.PipelineType == "StableDiffusionImg2ImgPipeline") or (request.IMG2IMG and request.PipelineType == ""):
                if fromSingleFile:
@ -188,7 +196,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                                                                             torch_dtype=torchType)
            ## img2vid
            elif request.PipelineType == "StableVideoDiffusionPipeline":
-                self.img2vid=True
+                self.img2vid = True
                self.pipe = StableVideoDiffusionPipeline.from_pretrained(
                    request.Model, torch_dtype=torchType, variant=variant
                )
@ -211,7 +219,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                self.pipe = DiffusionPipeline.from_pretrained(request.Model,
                                                              torch_dtype=torchType)
            elif request.PipelineType == "VideoDiffusionPipeline":
-                self.txt2vid=True
+                self.txt2vid = True
                self.pipe = DiffusionPipeline.from_pretrained(request.Model,
                                                              torch_dtype=torchType)
            elif request.PipelineType == "StableDiffusionXLPipeline":
@ -249,13 +257,12 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):

            if COMPEL:
                self.compel = Compel(
-                    tokenizer=[self.pipe.tokenizer, self.pipe.tokenizer_2 ], 
+                    tokenizer=[self.pipe.tokenizer, self.pipe.tokenizer_2],
                    text_encoder=[self.pipe.text_encoder, self.pipe.text_encoder_2],
                    returned_embeddings_type=ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NON_NORMALIZED,
                    requires_pooled=[False, True]
                )

-
            if request.ControlNet:
                self.controlnet = ControlNetModel.from_pretrained(
                    request.ControlNet, torch_dtype=torchType, variant=variant
@ -263,13 +270,6 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                self.pipe.controlnet = self.controlnet
            else:
                self.controlnet = None
-
-            if request.CUDA:
-                self.pipe.to('cuda')
-                if self.controlnet:
-                    self.controlnet.to('cuda')
-            if XPU:
-                self.pipe = self.pipe.to("xpu")
            # Assume directory from request.ModelFile.
            # Only if request.LoraAdapter it's not an absolute path
            if request.LoraAdapter and request.ModelFile != "" and not os.path.isabs(request.LoraAdapter) and request.LoraAdapter:
@ -282,10 +282,17 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
            if request.LoraAdapter:
                # Check if its a local file and not a directory ( we load lora differently for a safetensor file )
                if os.path.exists(request.LoraAdapter) and not os.path.isdir(request.LoraAdapter):
-                    self.load_lora_weights(request.LoraAdapter, 1, device, torchType)
+                    # self.load_lora_weights(request.LoraAdapter, 1, device, torchType)
+                    self.pipe.load_lora_weights(request.LoraAdapter)
                else:
                    self.pipe.unet.load_attn_procs(request.LoraAdapter)

+            if request.CUDA:
+                self.pipe.to('cuda')
+                if self.controlnet:
+                    self.controlnet.to('cuda')
+            if XPU:
+                self.pipe = self.pipe.to("xpu")
        except Exception as err:
            return backend_pb2.Result(success=False, message=f"Unexpected {err=}, {type(err)=}")
        # Implement your logic here for the LoadModel service
@ -372,7 +379,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
            options["image"] = pose_image

        if CLIPSKIP and self.clip_skip != 0:
-            options["clip_skip"]=self.clip_skip
+            options["clip_skip"] = self.clip_skip

        # Get the keys that we will build the args for our pipe for
        keys = options.keys()
@ -430,6 +437,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):

        return backend_pb2.Result(message="Media generated", success=True)

+
 def serve(address):
    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
    backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
@ -453,6 +461,7 @@ def serve(address):
    except KeyboardInterrupt:
        server.stop(0)

+
 if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Run the gRPC server.")
    parser.add_argument(
--- a/backend/python/diffusers/requirements-intel.txt
+++ b/backend/python/diffusers/requirements-intel.txt
@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchvision
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/diffusers/requirements.txt
+++ b/backend/python/diffusers/requirements.txt
@ -1,7 +1,9 @@
+setuptools
 accelerate
 compel
+peft
 diffusers
-grpcio==1.64.0
+grpcio==1.65.0
 opencv-python
 pillow
 protobuf
--- a/backend/python/exllama/requirements.txt
+++ b/backend/python/exllama/requirements.txt
@ -1,4 +1,4 @@
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 torch
 transformers
--- a/backend/python/exllama2/requirements.txt
+++ b/backend/python/exllama2/requirements.txt
@ -1,5 +1,5 @@
 accelerate
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 certifi
 torch
--- a/backend/python/mamba/requirements-install.txt
+++ b/backend/python/mamba/requirements-install.txt
@ -4,4 +4,4 @@
 packaging
 setuptools
 wheel
-torch==2.2.0
+torch==2.3.1
--- a/backend/python/mamba/requirements.txt
+++ b/backend/python/mamba/requirements.txt
@ -1,6 +1,6 @@
-causal-conv1d==1.2.0.post2
-mamba-ssm==1.2.0.post1
-grpcio==1.64.0
+causal-conv1d==1.4.0
+mamba-ssm==2.2.2
+grpcio==1.65.0
 protobuf
 certifi
 transformers
--- a/backend/python/openvoice/requirements-intel.txt
+++ b/backend/python/openvoice/requirements-intel.txt
@ -2,22 +2,22 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-grpcio==1.64.0
+grpcio==1.64.1
 protobuf
 librosa==0.9.1
-faster-whisper==0.9.0
+faster-whisper==1.0.3
 pydub==0.25.1
 wavmark==0.0.3
-numpy==1.22.0
+numpy==1.26.4
 eng_to_ipa==0.0.2
 inflect==7.0.0
 unidecode==1.3.7
-whisper-timestamped==1.14.2
+whisper-timestamped==1.15.4
 openai
 python-dotenv
 pypinyin==0.50.0
 cn2an==0.5.22
 jieba==0.42.1
-gradio==3.48.0
+gradio==4.38.1
 langid==1.1.6
 git+https://github.com/myshell-ai/MeloTTS.git
--- a/backend/python/openvoice/requirements.txt
+++ b/backend/python/openvoice/requirements.txt
@ -1,20 +1,20 @@
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
-librosa==0.9.1
-faster-whisper==0.9.0
+librosa
+faster-whisper
 pydub==0.25.1
 wavmark==0.0.3
-numpy==1.22.0
+numpy
 eng_to_ipa==0.0.2
-inflect==7.0.0
-unidecode==1.3.7
-whisper-timestamped==1.14.2
+inflect
+unidecode
+whisper-timestamped
 openai
 python-dotenv
-pypinyin==0.50.0
+pypinyin
 cn2an==0.5.22
 jieba==0.42.1
-gradio==3.48.0
+gradio
 langid==1.1.6
 git+https://github.com/myshell-ai/MeloTTS.git
 git+https://github.com/myshell-ai/OpenVoice.git
--- a/backend/python/parler-tts/requirements-intel.txt
+++ b/backend/python/parler-tts/requirements-intel.txt
@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/parler-tts/requirements.txt
+++ b/backend/python/parler-tts/requirements.txt
@ -1,5 +1,5 @@
 accelerate
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 torch
 git+https://github.com/huggingface/parler-tts.git@10016fb0300c0dc31a0fb70e26f3affee7b62f16
--- a/backend/python/petals/requirements-intel.txt
+++ b/backend/python/petals/requirements-intel.txt
@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/rerankers/requirements-intel.txt
+++ b/backend/python/rerankers/requirements-intel.txt
@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/rerankers/requirements.txt
+++ b/backend/python/rerankers/requirements.txt
@ -1,6 +1,6 @@
 accelerate
 rerankers[transformers]
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 certifi
 transformers
--- a/backend/python/sentencetransformers/requirements-intel.txt
+++ b/backend/python/sentencetransformers/requirements-intel.txt
@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/sentencetransformers/requirements.txt
+++ b/backend/python/sentencetransformers/requirements.txt
@ -1,6 +1,6 @@
 accelerate
-sentence-transformers==2.5.1
+sentence-transformers==3.0.1
 transformers
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 certifi
--- a/backend/python/transformers-musicgen/requirements-intel.txt
+++ b/backend/python/transformers-musicgen/requirements-intel.txt
@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/transformers-musicgen/requirements.txt
+++ b/backend/python/transformers-musicgen/requirements.txt
@ -1,7 +1,7 @@
 accelerate
 transformers
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 torch
-scipy==1.13.0
+scipy==1.14.0
 certifi
--- a/backend/python/transformers/requirements-intel.txt
+++ b/backend/python/transformers/requirements-intel.txt
@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/transformers/requirements.txt
+++ b/backend/python/transformers/requirements.txt
@ -1,9 +1,9 @@
 accelerate
 transformers
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 torch
 certifi
 intel-extension-for-transformers
 bitsandbytes
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/vall-e-x/requirements-intel.txt
+++ b/backend/python/vall-e-x/requirements-intel.txt
@ -3,4 +3,4 @@ intel-extension-for-pytorch
 torch
 torchaudio
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/vall-e-x/requirements.txt
+++ b/backend/python/vall-e-x/requirements.txt
@ -1,4 +1,4 @@
 accelerate
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 certifi
--- a/backend/python/vllm/requirements-intel.txt
+++ b/backend/python/vllm/requirements-intel.txt
@ -2,4 +2,4 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406
+setuptools==70.3.0 # https://github.com/mudler/LocalAI/issues/2406
--- a/backend/python/vllm/requirements.txt
+++ b/backend/python/vllm/requirements.txt
@ -1,6 +1,6 @@
 accelerate
 vllm
-grpcio==1.64.0
+grpcio==1.65.0
 protobuf
 certifi
 transformers
--- a/core/backend/options.go
+++ b/core/backend/options.go
@ -91,7 +91,7 @@ func gRPCModelOpts(c config.BackendConfig) *pb.ModelOptions {
 		Type:                 c.ModelType,
 		RopeFreqScale:        c.RopeFreqScale,
 		NUMA:                 c.NUMA,
-		Embeddings:           c.Embeddings,
+		Embeddings:           *c.Embeddings,
 		LowVRAM:              *c.LowVRAM,
 		NGPULayers:           int32(*c.NGPULayers),
 		MMap:                 *c.MMap,
--- a/core/cli/federated.go
+++ b/core/cli/federated.go
@ -2,20 +2,9 @@ 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 {
@ -25,106 +14,7 @@ type FederatedCLI struct {

 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)
-	}
+	fs := p2p.NewFederatedServer(f.Address, p2p.FederatedID, f.Peer2PeerToken)

-	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)
+	return fs.Start(context.Background())
 }
--- a/core/cli/run.go
+++ b/core/cli/run.go
@ -119,7 +119,7 @@ func (r *RunCMD) Run(ctx *cliContext.Context) error {
 		}

 		log.Info().Msg("Starting P2P server discovery...")
-		if err := p2p.ServiceDiscoverer(context.Background(), node, token, "", func() {
+		if err := p2p.ServiceDiscoverer(context.Background(), node, token, "", func(serviceID string, node p2p.NodeData) {
 			var tunnelAddresses []string
 			for _, v := range p2p.GetAvailableNodes("") {
 				if v.IsOnline() {
--- a/core/config/backend_config.go
+++ b/core/config/backend_config.go
@ -32,7 +32,7 @@ type BackendConfig struct {
 	Threads        *int              `yaml:"threads"`
 	Debug          *bool             `yaml:"debug"`
 	Roles          map[string]string `yaml:"roles"`
-	Embeddings     bool              `yaml:"embeddings"`
+	Embeddings     *bool             `yaml:"embeddings"`
 	Backend        string            `yaml:"backend"`
 	TemplateConfig TemplateConfig    `yaml:"template"`

@ -338,6 +338,10 @@ func (cfg *BackendConfig) SetDefaults(opts ...ConfigLoaderOption) {
 		cfg.LowVRAM = &falseV
 	}

+	if cfg.Embeddings == nil {
+		cfg.Embeddings = &falseV
+	}
+
 	// Value passed by the top level are treated as default (no implicit defaults)
 	// defaults are set by the user
 	if ctx == 0 {
--- a/core/config/guesser.go
+++ b/core/config/guesser.go
@ -20,6 +20,7 @@ const (
 	ChatML
 	Mistral03
 	Gemma
+	DeepSeek2
 )

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

 	switch {
+	case deepseek2:
+		return DeepSeek2
 	case gemma:
 		return Gemma
 	case llama3:
--- a/core/gallery/gallery.go
+++ b/core/gallery/gallery.go
@ -7,9 +7,10 @@ import (
 	"path/filepath"
 	"strings"

-	"github.com/imdario/mergo"
+	"dario.cat/mergo"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/pkg/downloader"
+	"github.com/mudler/LocalAI/pkg/utils"
 	"github.com/rs/zerolog/log"
 	"gopkg.in/yaml.v2"
 )
@ -189,6 +190,12 @@ func DeleteModelFromSystem(basePath string, name string, additionalFiles []strin

 	galleryFile := filepath.Join(basePath, galleryFileName(name))

+	for _, f := range []string{configFile, galleryFile} {
+		if err := utils.VerifyPath(f, basePath); err != nil {
+			return fmt.Errorf("failed to verify path %s: %w", f, err)
+		}
+	}
+
 	var err error
 	// Delete all the files associated to the model
 	// read the model config
--- a/core/gallery/models.go
+++ b/core/gallery/models.go
@ -6,7 +6,7 @@ import (
 	"os"
 	"path/filepath"

-	"github.com/imdario/mergo"
+	"dario.cat/mergo"
 	lconfig "github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/pkg/downloader"
 	"github.com/mudler/LocalAI/pkg/utils"
--- a/core/http/endpoints/jina/rerank.go
+++ b/core/http/endpoints/jina/rerank.go
@ -12,6 +12,11 @@ import (
 	"github.com/rs/zerolog/log"
 )

+// JINARerankEndpoint acts like the Jina reranker endpoint (https://jina.ai/reranker/)
+// @Summary Reranks a list of phrases by relevance to a given text query.
+// @Param request body schema.JINARerankRequest true "query params"
+// @Success 200 {object} schema.JINARerankResponse "Response"
+// @Router /v1/rerank [post]
 func JINARerankEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		req := new(schema.JINARerankRequest)
--- a/core/http/endpoints/localai/backend_monitor.go
+++ b/core/http/endpoints/localai/backend_monitor.go
@ -6,6 +6,11 @@ import (
 	"github.com/mudler/LocalAI/core/services"
 )

+// BackendMonitorEndpoint returns the status of the specified backend
+// @Summary Backend monitor endpoint
+// @Param request body schema.BackendMonitorRequest true "Backend statistics request"
+// @Success 200 {object} proto.StatusResponse "Response"
+// @Router /backend/monitor [get]
 func BackendMonitorEndpoint(bm *services.BackendMonitorService) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {

@ -23,6 +28,10 @@ func BackendMonitorEndpoint(bm *services.BackendMonitorService) func(c *fiber.Ct
 	}
 }

+// BackendMonitorEndpoint shuts down the specified backend
+// @Summary Backend monitor endpoint
+// @Param request body schema.BackendMonitorRequest true "Backend statistics request"
+// @Router /backend/shutdown [post]
 func BackendShutdownEndpoint(bm *services.BackendMonitorService) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(schema.BackendMonitorRequest)
--- a/core/http/endpoints/localai/gallery.go
+++ b/core/http/endpoints/localai/gallery.go
@ -9,6 +9,7 @@ import (
 	"github.com/google/uuid"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/rs/zerolog/log"
 )
@ -33,6 +34,10 @@ func CreateModelGalleryEndpointService(galleries []config.Gallery, modelPath str
 	}
 }

+// GetOpStatusEndpoint returns the job status
+// @Summary Returns the job status
+// @Success 200 {object} gallery.GalleryOpStatus "Response"
+// @Router /models/jobs/{uuid} [get]
 func (mgs *ModelGalleryEndpointService) GetOpStatusEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		status := mgs.galleryApplier.GetStatus(c.Params("uuid"))
@ -43,12 +48,21 @@ func (mgs *ModelGalleryEndpointService) GetOpStatusEndpoint() func(c *fiber.Ctx)
 	}
 }

+// GetAllStatusEndpoint returns all the jobs status progress
+// @Summary Returns all the jobs status progress
+// @Success 200 {object} map[string]gallery.GalleryOpStatus "Response"
+// @Router /models/jobs [get]
 func (mgs *ModelGalleryEndpointService) GetAllStatusEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		return c.JSON(mgs.galleryApplier.GetAllStatus())
 	}
 }

+// ApplyModelGalleryEndpoint installs a new model to a LocalAI instance from the model gallery
+// @Summary Install models to LocalAI.
+// @Param request body GalleryModel true "query params"
+// @Success 200 {object} schema.GalleryResponse "Response"
+// @Router /models/apply [post]
 func (mgs *ModelGalleryEndpointService) ApplyModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(GalleryModel)
@ -68,13 +82,15 @@ func (mgs *ModelGalleryEndpointService) ApplyModelGalleryEndpoint() func(c *fibe
 			Galleries:        mgs.galleries,
 			ConfigURL:        input.ConfigURL,
 		}
-		return c.JSON(struct {
-			ID        string `json:"uuid"`
-			StatusURL string `json:"status"`
-		}{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
+		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
 	}
 }

+// DeleteModelGalleryEndpoint lets delete models from a LocalAI instance
+// @Summary delete models to LocalAI.
+// @Param name	path string	true	"Model name"
+// @Success 200 {object} schema.GalleryResponse "Response"
+// @Router /models/delete/{name} [post]
 func (mgs *ModelGalleryEndpointService) DeleteModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		modelName := c.Params("name")
@ -89,13 +105,14 @@ func (mgs *ModelGalleryEndpointService) DeleteModelGalleryEndpoint() func(c *fib
 			return err
 		}

-		return c.JSON(struct {
-			ID        string `json:"uuid"`
-			StatusURL string `json:"status"`
-		}{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
+		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
 	}
 }

+// ListModelFromGalleryEndpoint list the available models for installation from the active galleries
+// @Summary List installable models.
+// @Success 200 {object} []gallery.GalleryModel "Response"
+// @Router /models/available [get]
 func (mgs *ModelGalleryEndpointService) ListModelFromGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		log.Debug().Msgf("Listing models from galleries: %+v", mgs.galleries)
@ -116,6 +133,10 @@ func (mgs *ModelGalleryEndpointService) ListModelFromGalleryEndpoint() func(c *f
 	}
 }

+// ListModelGalleriesEndpoint list the available galleries configured in LocalAI
+// @Summary List all Galleries
+// @Success 200 {object} []config.Gallery "Response"
+// @Router /models/galleries [get]
 // NOTE: This is different (and much simpler!) than above! This JUST lists the model galleries that have been loaded, not their contents!
 func (mgs *ModelGalleryEndpointService) ListModelGalleriesEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
@ -128,6 +149,11 @@ func (mgs *ModelGalleryEndpointService) ListModelGalleriesEndpoint() func(c *fib
 	}
 }

+// AddModelGalleryEndpoint adds a gallery in LocalAI
+// @Summary Adds a gallery in LocalAI
+// @Param request body config.Gallery true "Gallery details"
+// @Success 200 {object} []config.Gallery "Response"
+// @Router /models/galleries [post]
 func (mgs *ModelGalleryEndpointService) AddModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(config.Gallery)
@ -150,6 +176,11 @@ func (mgs *ModelGalleryEndpointService) AddModelGalleryEndpoint() func(c *fiber.
 	}
 }

+// RemoveModelGalleryEndpoint remove a gallery in LocalAI
+// @Summary removes a gallery from LocalAI
+// @Param request body config.Gallery true "Gallery details"
+// @Success 200 {object} []config.Gallery "Response"
+// @Router /models/galleries [delete]
 func (mgs *ModelGalleryEndpointService) RemoveModelGalleryEndpoint() func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		input := new(config.Gallery)
@ -165,6 +196,10 @@ func (mgs *ModelGalleryEndpointService) RemoveModelGalleryEndpoint() func(c *fib
 		mgs.galleries = slices.DeleteFunc(mgs.galleries, func(gallery config.Gallery) bool {
 			return gallery.Name == input.Name
 		})
-		return c.Send(nil)
+		dat, err := json.Marshal(mgs.galleries)
+		if err != nil {
+			return err
+		}
+		return c.Send(dat)
 	}
 }
--- a/core/http/endpoints/localai/metrics.go
+++ b/core/http/endpoints/localai/metrics.go
@ -9,8 +9,11 @@ import (
 	"github.com/prometheus/client_golang/prometheus/promhttp"
 )

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

--- a/core/http/endpoints/localai/p2p.go
+++ b/core/http/endpoints/localai/p2p.go
@ -0,0 +1,28 @@
+package localai
+
+import (
+	"github.com/gofiber/fiber/v2"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/p2p"
+	"github.com/mudler/LocalAI/core/schema"
+)
+
+// ShowP2PNodes returns the P2P Nodes
+// @Summary Returns available P2P nodes
+// @Success 200 {object} []schema.P2PNodesResponse "Response"
+// @Router /api/p2p [get]
+func ShowP2PNodes(c *fiber.Ctx) error {
+	// Render index
+	return c.JSON(schema.P2PNodesResponse{
+		Nodes:          p2p.GetAvailableNodes(""),
+		FederatedNodes: p2p.GetAvailableNodes(p2p.FederatedID),
+	})
+}
+
+// ShowP2PToken returns the P2P token
+// @Summary Show the P2P token
+// @Success 200 {string} string	 "Response"
+// @Router /api/p2p/token [get]
+func ShowP2PToken(appConfig *config.ApplicationConfig) func(*fiber.Ctx) error {
+	return func(c *fiber.Ctx) error { return c.Send([]byte(appConfig.P2PToken)) }
+}
--- a/core/http/endpoints/openai/assistant.go
+++ b/core/http/endpoints/openai/assistant.go
@ -11,6 +11,7 @@ import (

 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/core/services"
 	model "github.com/mudler/LocalAI/pkg/model"
 	"github.com/mudler/LocalAI/pkg/utils"
@ -125,6 +126,14 @@ func generateRandomID() int64 {
 	return currentId
 }

+// ListAssistantsEndpoint is the OpenAI Assistant API endpoint to list assistents https://platform.openai.com/docs/api-reference/assistants/listAssistants
+// @Summary List available assistents
+// @Param limit query int false "Limit the number of assistants returned"
+// @Param order query string false "Order of assistants returned"
+// @Param after query string false "Return assistants created after the given ID"
+// @Param before query string false "Return assistants created before the given ID"
+// @Success 200 {object} []Assistant "Response"
+// @Router /v1/assistants [get]
 func ListAssistantsEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		// Because we're altering the existing assistants list we should just duplicate it for now.
@ -230,13 +239,11 @@ func modelExists(cl *config.BackendConfigLoader, ml *model.ModelLoader, modelNam
 	return
 }

+// DeleteAssistantEndpoint is the OpenAI Assistant API endpoint to delete assistents https://platform.openai.com/docs/api-reference/assistants/deleteAssistant
+// @Summary Delete assistents
+// @Success 200 {object} schema.DeleteAssistantResponse "Response"
+// @Router /v1/assistants/{assistant_id} [delete]
 func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
-	type DeleteAssistantResponse struct {
-		ID      string `json:"id"`
-		Object  string `json:"object"`
-		Deleted bool   `json:"deleted"`
-	}
-
 	return func(c *fiber.Ctx) error {
 		assistantID := c.Params("assistant_id")
 		if assistantID == "" {
@ -247,7 +254,7 @@ func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoad
 			if assistant.ID == assistantID {
 				Assistants = append(Assistants[:i], Assistants[i+1:]...)
 				utils.SaveConfig(appConfig.ConfigsDir, AssistantsConfigFile, Assistants)
-				return c.Status(fiber.StatusOK).JSON(DeleteAssistantResponse{
+				return c.Status(fiber.StatusOK).JSON(schema.DeleteAssistantResponse{
 					ID:      assistantID,
 					Object:  "assistant.deleted",
 					Deleted: true,
@ -256,7 +263,7 @@ func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoad
 		}

 		log.Warn().Msgf("Unable to find assistant %s for deletion", assistantID)
-		return c.Status(fiber.StatusNotFound).JSON(DeleteAssistantResponse{
+		return c.Status(fiber.StatusNotFound).JSON(schema.DeleteAssistantResponse{
 			ID:      assistantID,
 			Object:  "assistant.deleted",
 			Deleted: false,
@ -264,6 +271,10 @@ func DeleteAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoad
 	}
 }

+// GetAssistantEndpoint is the OpenAI Assistant API endpoint to get assistents https://platform.openai.com/docs/api-reference/assistants/getAssistant
+// @Summary Get assistent data
+// @Success 200 {object} Assistant "Response"
+// @Router /v1/assistants/{assistant_id} [get]
 func GetAssistantEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		assistantID := c.Params("assistant_id")
@ -293,19 +304,9 @@ var (
 	AssistantsFileConfigFile = "assistantsFile.json"
 )

-type AssistantFileRequest struct {
-	FileID string `json:"file_id"`
-}
-
-type DeleteAssistantFileResponse struct {
-	ID      string `json:"id"`
-	Object  string `json:"object"`
-	Deleted bool   `json:"deleted"`
-}
-
 func CreateAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		request := new(AssistantFileRequest)
+		request := new(schema.AssistantFileRequest)
 		if err := c.BodyParser(request); err != nil {
 			return c.Status(fiber.StatusBadRequest).JSON(fiber.Map{"error": "Cannot parse JSON"})
 		}
@ -346,7 +347,7 @@ func CreateAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model

 func ListAssistantFilesEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	type ListAssistantFiles struct {
-		Data   []File
+		Data   []schema.File
 		Object string
 	}

@ -464,7 +465,7 @@ func DeleteAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 							// Remove the file from the assistantFiles slice
 							AssistantFiles = append(AssistantFiles[:i], AssistantFiles[i+1:]...)
 							utils.SaveConfig(appConfig.ConfigsDir, AssistantsFileConfigFile, AssistantFiles)
-							return c.Status(fiber.StatusOK).JSON(DeleteAssistantFileResponse{
+							return c.Status(fiber.StatusOK).JSON(schema.DeleteAssistantFileResponse{
 								ID:      fileId,
 								Object:  "assistant.file.deleted",
 								Deleted: true,
@ -480,7 +481,7 @@ func DeleteAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 						AssistantFiles = append(AssistantFiles[:i], AssistantFiles[i+1:]...)
 						utils.SaveConfig(appConfig.ConfigsDir, AssistantsFileConfigFile, AssistantFiles)

-						return c.Status(fiber.StatusNotFound).JSON(DeleteAssistantFileResponse{
+						return c.Status(fiber.StatusNotFound).JSON(schema.DeleteAssistantFileResponse{
 							ID:      fileId,
 							Object:  "assistant.file.deleted",
 							Deleted: true,
@ -491,7 +492,7 @@ func DeleteAssistantFileEndpoint(cl *config.BackendConfigLoader, ml *model.Model
 		}
 		log.Warn().Msgf("Unable to find assistant: %s", assistantID)

-		return c.Status(fiber.StatusNotFound).JSON(DeleteAssistantFileResponse{
+		return c.Status(fiber.StatusNotFound).JSON(schema.DeleteAssistantFileResponse{
 			ID:      fileId,
 			Object:  "assistant.file.deleted",
 			Deleted: false,
--- a/core/http/endpoints/openai/assistant_test.go
+++ b/core/http/endpoints/openai/assistant_test.go
@ -14,6 +14,7 @@ import (

 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/stretchr/testify/assert"
 )
@ -26,7 +27,7 @@ type MockLoader struct {

 func tearDown() func() {
 	return func() {
-		UploadedFiles = []File{}
+		UploadedFiles = []schema.File{}
 		Assistants = []Assistant{}
 		AssistantFiles = []AssistantFile{}
 		_ = os.Remove(filepath.Join(configsDir, AssistantsConfigFile))
@ -294,7 +295,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)

-		afr := AssistantFileRequest{FileID: file.ID}
+		afr := schema.AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)

 		assert.NoError(t, err)
@ -305,7 +306,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)

-		afr := AssistantFileRequest{FileID: file.ID}
+		afr := schema.AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)
 		assert.NoError(t, err)

@ -316,7 +317,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)

-		afr := AssistantFileRequest{FileID: file.ID}
+		afr := schema.AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)
 		assert.NoError(t, err)
 		t.Cleanup(cleanupAssistantFile(t, app, af.ID, af.AssistantID))
@ -338,7 +339,7 @@ func TestAssistantEndpoints(t *testing.T) {
 		file, assistant, err := createFileAndAssistant(t, app, appConfig)
 		assert.NoError(t, err)

-		afr := AssistantFileRequest{FileID: file.ID}
+		afr := schema.AssistantFileRequest{FileID: file.ID}
 		af, _, err := createAssistantFile(app, afr, assistant.ID)
 		assert.NoError(t, err)

@ -349,7 +350,7 @@ func TestAssistantEndpoints(t *testing.T) {

 }

-func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationConfig) (File, Assistant, error) {
+func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationConfig) (schema.File, Assistant, error) {
 	ar := &AssistantRequest{
 		Model:        "ggml-gpt4all-j",
 		Name:         "3.5-turbo",
@ -362,7 +363,7 @@ func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationC

 	assistant, _, err := createAssistant(app, *ar)
 	if err != nil {
-		return File{}, Assistant{}, err
+		return schema.File{}, Assistant{}, err
 	}
 	t.Cleanup(cleanupAllAssistants(t, app, []string{assistant.ID}))

@ -374,7 +375,7 @@ func createFileAndAssistant(t *testing.T, app *fiber.App, o *config.ApplicationC
 	return file, assistant, nil
 }

-func createAssistantFile(app *fiber.App, afr AssistantFileRequest, assistantId string) (AssistantFile, *http.Response, error) {
+func createAssistantFile(app *fiber.App, afr schema.AssistantFileRequest, assistantId string) (AssistantFile, *http.Response, error) {
 	afrJson, err := json.Marshal(afr)
 	if err != nil {
 		return AssistantFile{}, nil, err
@ -451,7 +452,7 @@ func cleanupAssistantFile(t *testing.T, app *fiber.App, fileId, assistantId stri
 		resp, err := app.Test(request)
 		assert.NoError(t, err)

-		var dafr DeleteAssistantFileResponse
+		var dafr schema.DeleteAssistantFileResponse
 		err = json.NewDecoder(resp.Body).Decode(&dafr)
 		assert.NoError(t, err)
 		assert.True(t, dafr.Deleted)
--- a/core/http/endpoints/openai/chat.go
+++ b/core/http/endpoints/openai/chat.go
@ -225,18 +225,10 @@ func ChatEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, startup
 			}

 			// Update input grammar
-			// Handle if we should return "name" instead of "functions"
-			if config.FunctionsConfig.FunctionName {
-				jsStruct := funcs.ToJSONNameStructure()
+			jsStruct := funcs.ToJSONStructure(config.FunctionsConfig.FunctionNameKey, config.FunctionsConfig.FunctionNameKey)
 			config.Grammar = jsStruct.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
-			} else {
-				jsStruct := funcs.ToJSONFunctionStructure()
-				config.Grammar = jsStruct.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
-			}
 		case input.JSONFunctionGrammarObject != nil:
 			config.Grammar = input.JSONFunctionGrammarObject.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
-		case input.JSONFunctionGrammarObjectName != nil:
-			config.Grammar = input.JSONFunctionGrammarObjectName.Grammar(config.FunctionsConfig.GrammarConfig.Options()...)
 		default:
 			// Force picking one of the functions by the request
 			if config.FunctionToCall() != "" {
--- a/core/http/endpoints/openai/edit.go
+++ b/core/http/endpoints/openai/edit.go
@ -16,6 +16,11 @@ import (
 	"github.com/rs/zerolog/log"
 )

+// EditEndpoint is the OpenAI edit API endpoint
+// @Summary OpenAI edit endpoint
+// @Param request body schema.OpenAIRequest true "query params"
+// @Success 200 {object} schema.OpenAIResponse "Response"
+// @Router /v1/edits [post]
 func EditEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 		modelFile, input, err := readRequest(c, cl, ml, appConfig, true)
--- a/core/http/endpoints/openai/files.go
+++ b/core/http/endpoints/openai/files.go
@ -9,25 +9,16 @@ import (
 	"time"

 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"

 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/pkg/utils"
 )

-var UploadedFiles []File
+var UploadedFiles []schema.File

 const UploadedFilesFile = "uploadedFiles.json"

-// File represents the structure of a file object from the OpenAI API.
-type File struct {
-	ID        string    `json:"id"`         // Unique identifier for the file
-	Object    string    `json:"object"`     // Type of the object (e.g., "file")
-	Bytes     int       `json:"bytes"`      // Size of the file in bytes
-	CreatedAt time.Time `json:"created_at"` // The time at which the file was created
-	Filename  string    `json:"filename"`   // The name of the file
-	Purpose   string    `json:"purpose"`    // The purpose of the file (e.g., "fine-tune", "classifications", etc.)
-}
-
 // UploadFilesEndpoint https://platform.openai.com/docs/api-reference/files/create
 func UploadFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
@ -61,7 +52,7 @@ func UploadFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Appli
 			return c.Status(fiber.StatusInternalServerError).SendString("Failed to save file: " + err.Error())
 		}

-		f := File{
+		f := schema.File{
 			ID:        fmt.Sprintf("file-%d", getNextFileId()),
 			Object:    "file",
 			Bytes:     int(file.Size),
@ -84,14 +75,13 @@ func getNextFileId() int64 {
 }

 // ListFilesEndpoint https://platform.openai.com/docs/api-reference/files/list
+// @Summary List files.
+// @Success 200 {object} schema.ListFiles "Response"
+// @Router /v1/files [get]
 func ListFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
-	type ListFiles struct {
-		Data   []File
-		Object string
-	}

 	return func(c *fiber.Ctx) error {
-		var listFiles ListFiles
+		var listFiles schema.ListFiles

 		purpose := c.Query("purpose")
 		if purpose == "" {
@ -108,7 +98,7 @@ func ListFilesEndpoint(cm *config.BackendConfigLoader, appConfig *config.Applica
 	}
 }

-func getFileFromRequest(c *fiber.Ctx) (*File, error) {
+func getFileFromRequest(c *fiber.Ctx) (*schema.File, error) {
 	id := c.Params("file_id")
 	if id == "" {
 		return nil, fmt.Errorf("file_id parameter is required")
@ -125,7 +115,7 @@ func getFileFromRequest(c *fiber.Ctx) (*File, error) {

 // 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"
+// @Success 200 {object} schema.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 {
--- a/core/http/endpoints/openai/files_test.go
+++ b/core/http/endpoints/openai/files_test.go
@ -14,6 +14,7 @@ import (
 	"github.com/rs/zerolog/log"

 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"

 	"github.com/gofiber/fiber/v2"
 	utils2 "github.com/mudler/LocalAI/pkg/utils"
@ -22,11 +23,6 @@ import (
 	"testing"
 )

-type ListFiles struct {
-	Data   []File
-	Object string
-}
-
 func startUpApp() (app *fiber.App, option *config.ApplicationConfig, loader *config.BackendConfigLoader) {
 	// Preparing the mocked objects
 	loader = &config.BackendConfigLoader{}
@ -159,7 +155,7 @@ func TestUploadFileExceedSizeLimit(t *testing.T) {
 		resp, _ := app.Test(req)
 		assert.Equal(t, 200, resp.StatusCode)

-		var listFiles ListFiles
+		var listFiles schema.ListFiles
 		if err := json.Unmarshal(bodyToByteArray(resp, t), &listFiles); err != nil {
 			t.Errorf("Failed to decode response: %v", err)
 			return
@ -201,7 +197,7 @@ func CallFilesUploadEndpoint(t *testing.T, app *fiber.App, fileName, tag, purpos
 	return app.Test(req)
 }

-func CallFilesUploadEndpointWithCleanup(t *testing.T, app *fiber.App, fileName, tag, purpose string, fileSize int, appConfig *config.ApplicationConfig) File {
+func CallFilesUploadEndpointWithCleanup(t *testing.T, app *fiber.App, fileName, tag, purpose string, fileSize int, appConfig *config.ApplicationConfig) schema.File {
 	// Create a file that exceeds the limit
 	testName := strings.Split(t.Name(), "/")[1]
 	file := createTestFile(t, testName+"-"+fileName, fileSize, appConfig)
@ -280,8 +276,8 @@ func bodyToByteArray(resp *http.Response, t *testing.T) []byte {
 	return bodyBytes
 }

-func responseToFile(t *testing.T, resp *http.Response) File {
-	var file File
+func responseToFile(t *testing.T, resp *http.Response) schema.File {
+	var file schema.File
 	responseToString := bodyToString(resp, t)

 	err := json.NewDecoder(strings.NewReader(responseToString)).Decode(&file)
@ -292,8 +288,8 @@ func responseToFile(t *testing.T, resp *http.Response) File {
 	return file
 }

-func responseToListFile(t *testing.T, resp *http.Response) ListFiles {
-	var listFiles ListFiles
+func responseToListFile(t *testing.T, resp *http.Response) schema.ListFiles {
+	var listFiles schema.ListFiles
 	responseToString := bodyToString(resp, t)

 	err := json.NewDecoder(strings.NewReader(responseToString)).Decode(&listFiles)
--- a/core/http/routes/localai.go
+++ b/core/http/routes/localai.go
@ -59,16 +59,8 @@ func RegisterLocalAIRoutes(app *fiber.App,

 	// p2p
 	if p2p.IsP2PEnabled() {
-		app.Get("/api/p2p", auth, func(c *fiber.Ctx) error {
-			// Render index
-			return c.JSON(map[string]interface{}{
-				"Nodes":          p2p.GetAvailableNodes(""),
-				"FederatedNodes": p2p.GetAvailableNodes(p2p.FederatedID),
-			})
-		})
-		app.Get("/api/p2p/token", auth, func(c *fiber.Ctx) error {
-			return c.Send([]byte(appConfig.P2PToken))
-		})
+		app.Get("/api/p2p", auth, localai.ShowP2PNodes)
+		app.Get("/api/p2p/token", auth, localai.ShowP2PToken(appConfig))
 	}

 	app.Get("/version", auth, func(c *fiber.Ctx) error {
--- a/core/p2p/federated.go
+++ b/core/p2p/federated.go
@ -0,0 +1,15 @@
+package p2p
+
+const FederatedID = "federated"
+
+type FederatedServer struct {
+	listenAddr, service, p2ptoken string
+}
+
+func NewFederatedServer(listenAddr, service, p2pToken string) *FederatedServer {
+	return &FederatedServer{
+		listenAddr: listenAddr,
+		service:    service,
+		p2ptoken:   p2pToken,
+	}
+}
--- a/core/p2p/federated_server.go
+++ b/core/p2p/federated_server.go
@ -0,0 +1,127 @@
+//go:build p2p
+// +build p2p
+
+package p2p
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"net"
+	"time"
+
+	"math/rand/v2"
+
+	"github.com/mudler/edgevpn/pkg/node"
+	"github.com/mudler/edgevpn/pkg/protocol"
+	"github.com/mudler/edgevpn/pkg/types"
+	"github.com/rs/zerolog/log"
+)
+
+func (f *FederatedServer) Start(ctx context.Context) error {
+
+	n, err := NewNode(f.p2ptoken)
+	if err != nil {
+		return fmt.Errorf("creating a new node: %w", err)
+	}
+	err = n.Start(ctx)
+	if err != nil {
+		return fmt.Errorf("creating a new node: %w", err)
+	}
+
+	if err := ServiceDiscoverer(ctx, n, f.p2ptoken, f.service, func(servicesID string, tunnel NodeData) {
+		log.Debug().Msgf("Discovered node: %s", tunnel.ID)
+	}); err != nil {
+		return err
+	}
+
+	return f.proxy(ctx, n)
+}
+
+func (fs *FederatedServer) proxy(ctx context.Context, node *node.Node) error {
+
+	log.Info().Msgf("Allocating service '%s' on: %s", fs.service, fs.listenAddr)
+	// Open local port for listening
+	l, err := net.Listen("tcp", fs.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 GetAvailableNodes(fs.service) {
+					if v.IsOnline() {
+						tunnelAddresses = append(tunnelAddresses, v.TunnelAddress)
+					} else {
+						log.Info().Msgf("Node %s is offline", v.ID)
+					}
+				}
+
+				if len(tunnelAddresses) == 0 {
+					log.Error().Msg("No available nodes yet")
+					return
+				}
+
+				// 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())
+			}()
+		}
+	}
+
+}
--- a/core/p2p/node.go
+++ b/core/p2p/node.go
@ -6,7 +6,6 @@ import (
 )

 const defaultServicesID = "services_localai"
-const FederatedID = "federated"

 type NodeData struct {
 	Name          string
--- a/core/p2p/p2p.go
+++ b/core/p2p/p2p.go
@ -137,14 +137,9 @@ func allocateLocalService(ctx context.Context, node *node.Node, listenAddr, serv

 }

-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)
-}
-
 // This is the main of the server (which keeps the env variable updated)
 // This starts a goroutine that keeps LLAMACPP_GRPC_SERVERS updated with the discovered services
-func ServiceDiscoverer(ctx context.Context, n *node.Node, token, servicesID string, discoveryFunc func()) error {
+func ServiceDiscoverer(ctx context.Context, n *node.Node, token, servicesID string, discoveryFunc func(serviceID string, node NodeData)) error {
 	if servicesID == "" {
 		servicesID = defaultServicesID
 	}
@ -166,7 +161,7 @@ func ServiceDiscoverer(ctx context.Context, n *node.Node, token, servicesID stri
 			case tunnel := <-tunnels:
 				AddNode(servicesID, tunnel)
 				if discoveryFunc != nil {
-					discoveryFunc()
+					discoveryFunc(servicesID, tunnel)
 				}
 			}
 		}
@ -396,3 +391,8 @@ func newNodeOpts(token string) ([]node.Option, error) {

 	return nodeOpts, nil
 }
+
+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)
+}
--- a/core/p2p/p2p_disabled.go
+++ b/core/p2p/p2p_disabled.go
@ -14,7 +14,11 @@ func GenerateToken() string {
 	return "not implemented"
 }

-func ServiceDiscoverer(ctx context.Context, node *node.Node, token, servicesID string, fn func()) error {
+func (f *FederatedServer) Start(ctx context.Context) error {
+	return fmt.Errorf("not implemented")
+}
+
+func ServiceDiscoverer(ctx context.Context, node *node.Node, token, servicesID string, fn func(string, NodeData)) error {
 	return fmt.Errorf("not implemented")
 }

--- a/core/schema/localai.go
+++ b/core/schema/localai.go
@ -1,6 +1,7 @@
 package schema

 import (
+	"github.com/mudler/LocalAI/core/p2p"
 	gopsutil "github.com/shirou/gopsutil/v3/process"
 )

@ -14,6 +15,11 @@ type BackendMonitorResponse struct {
 	CPUPercent    float64
 }

+type GalleryResponse struct {
+	ID        string `json:"uuid"`
+	StatusURL string `json:"status"`
+}
+
 // @Description TTS request body
 type TTSRequest struct {
 	Model    string `json:"model" yaml:"model"` // model name or full path
@ -59,3 +65,8 @@ type StoresFindResponse struct {
 	Values       []string    `json:"values" yaml:"values"`
 	Similarities []float32   `json:"similarities" yaml:"similarities"`
 }
+
+type P2PNodesResponse struct {
+	Nodes          []p2p.NodeData `json:"nodes" yaml:"nodes"`
+	FederatedNodes []p2p.NodeData `json:"federated_nodes" yaml:"federated_nodes"`
+}
--- a/core/schema/openai.go
+++ b/core/schema/openai.go
@ -2,6 +2,7 @@ package schema

 import (
 	"context"
+	"time"

 	functions "github.com/mudler/LocalAI/pkg/functions"
 )
@ -99,6 +100,37 @@ type OpenAIModel struct {
 	Object string `json:"object"`
 }

+type DeleteAssistantResponse struct {
+	ID      string `json:"id"`
+	Object  string `json:"object"`
+	Deleted bool   `json:"deleted"`
+}
+
+// File represents the structure of a file object from the OpenAI API.
+type File struct {
+	ID        string    `json:"id"`         // Unique identifier for the file
+	Object    string    `json:"object"`     // Type of the object (e.g., "file")
+	Bytes     int       `json:"bytes"`      // Size of the file in bytes
+	CreatedAt time.Time `json:"created_at"` // The time at which the file was created
+	Filename  string    `json:"filename"`   // The name of the file
+	Purpose   string    `json:"purpose"`    // The purpose of the file (e.g., "fine-tune", "classifications", etc.)
+}
+
+type ListFiles struct {
+	Data   []File
+	Object string
+}
+
+type AssistantFileRequest struct {
+	FileID string `json:"file_id"`
+}
+
+type DeleteAssistantFileResponse struct {
+	ID      string `json:"id"`
+	Object  string `json:"object"`
+	Deleted bool   `json:"deleted"`
+}
+
 type ImageGenerationResponseFormat string

 type ChatCompletionResponseFormatType string
@ -147,8 +179,7 @@ type OpenAIRequest struct {
 	// A grammar to constrain the LLM output
 	Grammar string `json:"grammar" yaml:"grammar"`

-	JSONFunctionGrammarObject     *functions.JSONFunctionStructureFunction `json:"grammar_json_functions" yaml:"grammar_json_functions"`
-	JSONFunctionGrammarObjectName *functions.JSONFunctionStructureName     `json:"grammar_json_name" yaml:"grammar_json_name"`
+	JSONFunctionGrammarObject *functions.JSONFunctionStructure `json:"grammar_json_functions" yaml:"grammar_json_functions"`

 	Backend string `json:"backend" yaml:"backend"`

--- a/core/startup/config_file_watcher.go
+++ b/core/startup/config_file_watcher.go
@ -9,7 +9,7 @@ import (
 	"time"

 	"github.com/fsnotify/fsnotify"
-	"github.com/imdario/mergo"
+	"dario.cat/mergo"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/rs/zerolog/log"
 )
--- a/docs/content/docs/advanced/advanced-usage.md
+++ b/docs/content/docs/advanced/advanced-usage.md
@ -112,6 +112,8 @@ name: "" # Model name, used to identify the model in API calls.
 # Precision settings for the model, reducing precision can enhance performance on some hardware.
 f16: null # Whether to use 16-bit floating-point precision.

+embeddings: true # Enable embeddings for the model.
+
 # Concurrency settings for the application.
 threads: null # Number of threads to use for processing.

@ -150,7 +152,8 @@ function:
    replace_function_results: [] # Placeholder to replace function call results with arbitrary strings or patterns.
    replace_llm_results: [] # Replace language model results with arbitrary strings or patterns.
    capture_llm_results: [] # Capture language model results as text result, among JSON, in function calls. For instance, if a model returns a block for "thinking" and a block for "response", this will allow you to capture the thinking block.
-    return_name_in_function_response: false # Some models might prefer to use "name" rather then "function" when returning JSON data. This will allow to use "name" as a key in the JSON response.
+    function_name_key: "name"
+    function_arguments_key: "arguments"

 # Feature gating flags to enable experimental or optional features.
 feature_flags: {}
--- a/docs/content/docs/advanced/run-other-models.md
+++ b/docs/content/docs/advanced/run-other-models.md
@ -8,9 +8,9 @@ icon = "rocket_launch"

 ## Running other models

-> _Do you have already a model file? Skip to [Run models manually]({{%relref "docs/getting-started/manual" %}})_.
+> _Do you have already a model file? Skip to [Run models manually]({{%relref "docs/getting-started/models" %}})_.

-To load models into LocalAI, you can either [use models manually]({{%relref "docs/getting-started/manual" %}}) or configure LocalAI to pull the models from external sources, like Huggingface and configure the model.
+To load models into LocalAI, you can either [use models manually]({{%relref "docs/getting-started/models" %}}) or configure LocalAI to pull the models from external sources, like Huggingface and configure the model.

 To do that, you can point LocalAI to an URL to a YAML configuration file - however - LocalAI does also have some popular model configuration embedded in the binary as well. Below you can find a list of the models configuration that LocalAI has pre-built, see [Model customization]({{%relref "docs/getting-started/customize-model" %}}) on how to configure models from URLs.

--- a/docs/content/docs/faq.md
+++ b/docs/content/docs/faq.md
@ -16,6 +16,10 @@ Here are answers to some of the most common questions.

 Most gguf-based models should work, but newer models may require additions to the API. If a model doesn't work, please feel free to open up issues. However, be cautious about downloading models from the internet and directly onto your machine, as there may be security vulnerabilities in lama.cpp or ggml that could be maliciously exploited. Some models can be found on Hugging Face: https://huggingface.co/models?search=gguf, or models from gpt4all are compatible too: https://github.com/nomic-ai/gpt4all.

+### Benchmarking LocalAI and llama.cpp shows different results!
+
+LocalAI applies a set of defaults when loading models with the llama.cpp backend, one of these is mirostat sampling - while it achieves better results, it slows down the inference. You can disable this by setting `mirostat: 0` in the model config file. See also the advanced section ({{%relref "docs/advanced/advanced-usage" %}}) for more information and [this issue](https://github.com/mudler/LocalAI/issues/2780).
+
 ### What's the difference with Serge, or XXX?

 LocalAI is a multi-model solution that doesn't focus on a specific model type (e.g., llama.cpp or alpaca.cpp), and it handles all of these internally for faster inference,  easy to set up locally and deploy to Kubernetes.
--- a/docs/content/docs/getting-started/build.md
+++ b/docs/content/docs/getting-started/build.md
@ -55,8 +55,8 @@ apt install cmake golang libgrpc-dev make protobuf-compiler-grpc python3-grpc-to
 After you have golang installed and working, you can install the required binaries for compiling the golang protobuf components via the following commands

 ```bash
-go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.0
-go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@8ba23be9613c672d40ae261d2a1335d639bdd59b
+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

 ```

--- a/docs/content/docs/getting-started/models.md
+++ b/docs/content/docs/getting-started/models.md
@ -1,21 +1,69 @@
---
+++
+disableToc = false
+title = "Install and Run Models"
+weight = 4
+icon = "rocket_launch"
+++

-disableToc: false
-title: "Run models manually"
-weight: 5
-icon: "rocket_launch"
+To install models with LocalAI, you can:

---
+- Browse the Model Gallery from the Web Interface and install models with a couple of clicks. For more details, refer to the [Gallery Documentation]({{% relref "docs/features/model-gallery" %}}).
+- Specify a model from the LocalAI gallery during startup, e.g., `local-ai run <model_gallery_name>`.
+- Use a URI to specify a model file (e.g., `huggingface://...`, `oci://`, or `ollama://`) when starting LocalAI, e.g., `local-ai run huggingface://TheBloke/phi-2-GGUF/phi-2.Q8_0.gguf`.
+- Specify a URL to a model configuration file when starting LocalAI, e.g., `local-ai run https://gist.githubusercontent.com/.../phi-2.yaml`.
+- Manually install the models by copying the files into the models directory (`--models`).

-# Run Models Manually
+## Run and Install Models via the Gallery
+
+To run models available in the LocalAI gallery, you can use the WebUI or specify the model name when starting LocalAI. Models can be found in the gallery via the Web interface, the [model gallery](https://models.localai.io), or the CLI with: `local-ai models list`.
+
+To install a model from the gallery, use the model name as the URI. For example, to run LocalAI with the Hermes model, execute:
+
+```bash
+local-ai run hermes-2-theta-llama-3-8b
+```
+
+To install only the model, use:
+
+```bash
+local-ai models install hermes-2-theta-llama-3-8b
+```
+
+Note: The galleries available in LocalAI can be customized to point to a different URL or a local directory. For more information on how to setup your own gallery, see the [Gallery Documentation]({{% relref "docs/features/model-gallery" %}}).
+
+## Run Models via URI
+
+To run models via URI, specify a URI to a model file or a configuration file when starting LocalAI. Valid syntax includes:
+
+- `file://path/to/model`
+- `huggingface://repository_id/model_file` (e.g., `huggingface://TheBloke/phi-2-GGUF/phi-2.Q8_0.gguf`)
+- From OCIs: `oci://container_image:tag`, `ollama://model_id:tag`
+- From configuration files: `https://gist.githubusercontent.com/.../phi-2.yaml`
+
+Configuration files can be used to customize the model defaults and settings. For advanced configurations, refer to the [Customize Models section]({{% relref "docs/getting-started/customize-model" %}}).
+
+### Examples
+
+```bash
+# Start LocalAI with the phi-2 model
+local-ai run huggingface://TheBloke/phi-2-GGUF/phi-2.Q8_0.gguf
+# Install and run a model from the Ollama OCI registry
+local-ai run ollama://gemma:2b
+# Run a model from a configuration file
+local-ai run https://gist.githubusercontent.com/.../phi-2.yaml
+# Install and run a model from a standard OCI registry (e.g., Docker Hub)
+local-ai run oci://localai/phi-2:latest
+```
+
+## Run Models Manually

 Follow these steps to manually run models using LocalAI:

 1. **Prepare Your Model and Configuration Files**:
-   Ensure you have a model file and a configuration YAML file, if necessary. Customize model defaults and specific settings with a configuration file. For advanced configurations, refer to the [Advanced Documentation]({{% relref "docs/advanced" %}}).
+   Ensure you have a model file and, if necessary, a configuration YAML file. Customize model defaults and settings with a configuration file. For advanced configurations, refer to the [Advanced Documentation]({{% relref "docs/advanced" %}}).

 2. **GPU Acceleration**:
-   For instructions on GPU acceleration, visit the [GPU acceleration]({{% relref "docs/features/gpu-acceleration" %}}) page.
+   For instructions on GPU acceleration, visit the [GPU Acceleration]({{% relref "docs/features/gpu-acceleration" %}}) page.

 3. **Run LocalAI**:
   Choose one of the following methods to run LocalAI:
@ -160,5 +208,3 @@ For instructions on building LocalAI from source, see the [Build Section]({{% re
 {{< /tabs >}}

 For more model configurations, visit the [Examples Section](https://github.com/mudler/LocalAI/tree/master/examples/configurations).
-
---
--- a/docs/content/docs/getting-started/quickstart.md
+++ b/docs/content/docs/getting-started/quickstart.md
@ -38,13 +38,13 @@ For detailed instructions, see [Using container images]({{% relref "docs/getting

 ## Running LocalAI with All-in-One (AIO) Images

-> _Already have a model file? Skip to [Run models manually]({{% relref "docs/getting-started/manual" %}})_.
+> _Already have a model file? Skip to [Run models manually]({{% relref "docs/getting-started/models" %}})_.

 LocalAI's All-in-One (AIO) images are pre-configured with a set of models and backends to fully leverage almost all the features of LocalAI. If pre-configured models are not required, you can use the standard [images]({{% relref "docs/getting-started/container-images" %}}).

 These images are available for both CPU and GPU environments. AIO images are designed for ease of use and require no additional configuration.

-It is recommended to use AIO images if you prefer not to configure the models manually or via the web interface. For running specific models, refer to the [manual method]({{% relref "docs/getting-started/manual" %}}).
+It is recommended to use AIO images if you prefer not to configure the models manually or via the web interface. For running specific models, refer to the [manual method]({{% relref "docs/getting-started/models" %}}).

 The AIO images come pre-configured with the following features:
 - Text to Speech (TTS)
@ -66,5 +66,5 @@ Explore additional resources and community contributions:
 - [Run from Container images]({{% relref "docs/getting-started/container-images" %}})
 - [Examples to try from the CLI]({{% relref "docs/getting-started/try-it-out" %}})
 - [Build LocalAI and the container image]({{% relref "docs/getting-started/build" %}})
- [Run models manually]({{% relref "docs/getting-started/manual" %}})
+- [Run models manually]({{% relref "docs/getting-started/models" %}})
 - [Examples](https://github.com/mudler/LocalAI/tree/master/examples#examples)
--- a/docs/content/docs/getting-started/try-it-out.md
+++ b/docs/content/docs/getting-started/try-it-out.md
@ -17,10 +17,10 @@ After installation, install new models by navigating the model gallery, or by us
 To install models with the WebUI, see the [Models section]({{%relref "docs/features/model-gallery" %}}).
 With the CLI you can list the models with `local-ai models list` and install them with `local-ai models install <model-name>`.

-You can also [run models manually]({{%relref "docs/getting-started/manual" %}}) by copying files into the `models` directory.
+You can also [run models manually]({{%relref "docs/getting-started/models" %}}) by copying files into the `models` directory.
 {{% /alert %}}

-You can test out the API endpoints using `curl`, few examples are listed below. The models we are refering here (`gpt-4`, `gpt-4-vision-preview`, `tts-1`, `whisper-1`) are the default models that come with the AIO images - you can also use any other model you have installed.
+You can test out the API endpoints using `curl`, few examples are listed below. The models we are referring here (`gpt-4`, `gpt-4-vision-preview`, `tts-1`, `whisper-1`) are the default models that come with the AIO images - you can also use any other model you have installed.

 ### Text Generation

--- a/docs/themes/hugo-theme-relearn
+++ b/docs/themes/hugo-theme-relearn
@ -1 +1 @@
-Subproject commit c25bc2a27ab46649393ef7b310e14fff1311116d
+Subproject commit 1b2e139512106f8074ac7d4a884135d159720cc4
--- a/examples/chainlit/Dockerfile
+++ b/examples/chainlit/Dockerfile
@ -1,5 +1,5 @@
 # Use an official Python runtime as a parent image
-FROM harbor.home.sfxworks.net/docker/library/python:3.9-slim
+FROM python:3.12-slim

 # Set the working directory in the container
 WORKDIR /app
@ -7,8 +7,17 @@ WORKDIR /app
 # Copy the current directory contents into the container at /app
 COPY requirements.txt /app

+# Install c++ compiler
+RUN apt-get update \
+ && DEBIAN_FRONTEND=noninteractive apt-get install -y build-essential \
+ && apt-get clean \
+ && rm -rf /var/lib/apt/lists/*
+
 # Install any needed packages specified in requirements.txt
-RUN pip install -r requirements.txt
+RUN pip install --no-cache-dir -r requirements.txt \
+ && DEBIAN_FRONTEND=noninteractive apt-get remove -y build-essential \
+ && apt-get clean \
+ && rm -rf /var/lib/apt/lists/*

 COPY . /app

--- a/examples/chainlit/requirements.txt
+++ b/examples/chainlit/requirements.txt
@ -1,7 +1,6 @@
-llama_hub==0.0.41
-llama_index==0.8.55
-Requests==2.31.0
-weaviate_client==3.25.1
+llama_index==0.10.55
+requests==2.32.3
+weaviate_client==4.6.5
 transformers
 torch
 chainlit
--- a/examples/functions/Dockerfile
+++ b/examples/functions/Dockerfile
@ -1,5 +1,5 @@
-FROM python:3.10-bullseye
+FROM python:3.12-slim-bullseye
 COPY . /app
 WORKDIR /app
 RUN pip install --no-cache-dir -r requirements.txt
-ENTRYPOINT [ "python", "./functions-openai.py" ];
+ENTRYPOINT [ "python", "./functions-openai.py" ]
--- a/examples/functions/requirements.txt
+++ b/examples/functions/requirements.txt
@ -1,2 +1,2 @@
-langchain==0.1.0
-openai==0.27.8
+langchain==0.2.8
+openai==1.35.13
--- a/examples/k8sgpt/broken-pod.yaml
+++ b/examples/k8sgpt/broken-pod.yaml
@ -5,7 +5,7 @@ metadata:
 spec:
  containers:
    - name: broken-pod
-      image: nginx:1.a.b.c
+      image: nginx:1.27.0
      livenessProbe:
        httpGet:
          path: /
--- a/examples/k8sgpt/values.yaml
+++ b/examples/k8sgpt/values.yaml
@ -2,7 +2,7 @@ replicaCount: 1

 deployment:
  # https://quay.io/repository/go-skynet/local-ai?tab=tags
-  image: quay.io/go-skynet/local-ai:v1.23.0
+  image: quay.io/go-skynet/local-ai:v1.40.0
  env:
    threads: 4
    debug: "true"
--- a/examples/langchain-chroma/requirements.txt
+++ b/examples/langchain-chroma/requirements.txt
@ -1,4 +1,4 @@
-langchain==0.1.0
-openai==0.27.6
-chromadb==0.3.21
-llama-index==0.9.36
+langchain==0.2.8
+openai==1.35.13
+chromadb==0.5.4
+llama-index==0.10.55
--- a/examples/langchain/JS.Dockerfile
+++ b/examples/langchain/JS.Dockerfile
@ -1,4 +1,4 @@
-FROM node:latest
+FROM node:lts-alpine
 COPY ./langchainjs-localai-example /app
 WORKDIR /app
 RUN npm install
--- a/examples/langchain/PY.Dockerfile
+++ b/examples/langchain/PY.Dockerfile
@ -1,5 +1,5 @@
-FROM python:3.10-bullseye
+FROM python:3.12-bullseye
 COPY ./langchainpy-localai-example /app
 WORKDIR /app
 RUN pip install --no-cache-dir -r requirements.txt
-ENTRYPOINT [ "python", "./full_demo.py" ];
+ENTRYPOINT [ "python", "./full_demo.py" ]
--- a/examples/langchain/langchainpy-localai-example/requirements.txt
+++ b/examples/langchain/langchainpy-localai-example/requirements.txt
@ -1,32 +1,33 @@
-aiohttp==3.9.4
+aiohttp==3.9.5
 aiosignal==1.3.1
-async-timeout==4.0.2
-attrs==23.1.0
-certifi==2023.7.22
-charset-normalizer==3.1.0
+async-timeout==4.0.3
+attrs==23.2.0
+certifi==2024.7.4
+charset-normalizer==3.3.2
 colorama==0.4.6
-dataclasses-json==0.5.7
-debugpy==1.6.7
-frozenlist==1.3.3
-greenlet==2.0.2
+dataclasses-json==0.6.7
+debugpy==1.8.2
+frozenlist==1.4.1
+greenlet==3.0.3
 idna==3.7
-langchain==0.1.0
-marshmallow==3.19.0
+langchain==0.2.8
+langchain-community==0.2.7
+marshmallow==3.21.3
 marshmallow-enum==1.5.1
-multidict==6.0.4
+multidict==6.0.5
 mypy-extensions==1.0.0
-numexpr==2.8.4
-numpy==1.24.3
-openai==0.27.6
+numexpr==2.10.1
+numpy==1.26.4
+openai==1.35.13
 openapi-schema-pydantic==1.2.4
-packaging==23.1
-pydantic==1.10.13
-PyYAML==6.0
-requests==2.31.0
-SQLAlchemy==2.0.12
-tenacity==8.2.2
-tqdm==4.66.3
-typing-inspect==0.8.0
-typing_extensions==4.5.0
-urllib3==1.26.18
-yarl==1.9.2
+packaging>=23.2
+pydantic==2.8.2
+PyYAML==6.0.1
+requests==2.32.3
+SQLAlchemy==2.0.30
+tenacity==8.5.0
+tqdm==4.66.4
+typing-inspect==0.9.0
+typing_extensions==4.12.2
+urllib3==2.2.2
+yarl==1.9.4
--- a/examples/semantic-todo/go.mod
+++ b/examples/semantic-todo/go.mod
@ -1,10 +1,10 @@
 module semantic-todo

-go 1.21.6
+go 1.22

 require (
-	github.com/gdamore/tcell/v2 v2.7.1
-	github.com/rivo/tview v0.0.0-20240307173318-e804876934a1
+	github.com/gdamore/tcell/v2 v2.7.4
+	github.com/rivo/tview v0.0.0-20240524063012-037df494fb76
 )

 require (
--- a/Show More
+++ b/Show More