fixup: create piper libdir also when not built

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Merge branch 'master' into enable_gpu
2026-02-03 11:13:31 -05:00 · 2023-11-12 22:17:11 +01:00 · 2023-11-11 19:20:36 +01:00 · 2023-11-11 18:40:48 +01:00 · 2023-11-11 18:40:26 +01:00 · 2023-11-11 13:14:59 +01:00
86 changed files with 28847 additions and 1180 deletions
--- a/.env
+++ b/.env
@@ -66,4 +66,7 @@ MODELS_PATH=/models
 ### Python backends GRPC max workers
 ### Default number of workers for GRPC Python backends.
 ### This actually controls wether a backend can process multiple requests or not.
-# PYTHON_GRPC_MAX_WORKERS=1
+# PYTHON_GRPC_MAX_WORKERS=1
+
+### Define the number of parallel LLAMA.cpp workers (Defaults to 1)
+# LLAMACPP_PARALLEL=1
--- a/.github/workflows/image.yml
+++ b/.github/workflows/image.yml
@@ -14,7 +14,7 @@ concurrency:
  cancel-in-progress: true

 jobs:
-  docker:
+  image-build:
    strategy:
      matrix:
        include:
@@ -24,6 +24,11 @@ jobs:
            tag-latest: 'auto'
            tag-suffix: ''
            ffmpeg: ''
+          - build-type: ''
+            platforms: 'linux/amd64'
+            tag-latest: 'false'
+            tag-suffix: '-ffmpeg'
+            ffmpeg: 'true'
          - build-type: 'cublas'
            cuda-major-version: 11
            cuda-minor-version: 7
@@ -38,11 +43,6 @@ jobs:
            tag-latest: 'false'
            tag-suffix: '-cublas-cuda12'
            ffmpeg: ''
-          - build-type: ''
-            platforms: 'linux/amd64'
-            tag-latest: 'false'
-            tag-suffix: '-ffmpeg'
-            ffmpeg: 'true'
          - build-type: 'cublas'
            cuda-major-version: 11
            cuda-minor-version: 7
@@ -58,47 +58,54 @@ jobs:
            tag-suffix: '-cublas-cuda12-ffmpeg'
            ffmpeg: 'true'

-    runs-on: ubuntu-latest
+    runs-on: arc-runner-set 
    steps:
-      - name: Release space from worker
+      - name: Force Install GIT latest
        run: |
-          echo "Listing top largest packages"
-          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
-          head -n 30 <<< "${pkgs}"
-          echo
-          df -h
-          echo
-          sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
-          sudo apt-get remove --auto-remove android-sdk-platform-tools || true
-          sudo apt-get purge --auto-remove android-sdk-platform-tools || true
-          sudo rm -rf /usr/local/lib/android
-          sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
-          sudo rm -rf /usr/share/dotnet
-          sudo apt-get remove -y '^mono-.*' || true
-          sudo apt-get remove -y '^ghc-.*' || true
-          sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
-          sudo apt-get remove -y 'php.*' || true
-          sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
-          sudo apt-get remove -y '^google-.*' || true
-          sudo apt-get remove -y azure-cli || true
-          sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
-          sudo apt-get remove -y '^gfortran-.*' || true
-          sudo apt-get remove -y microsoft-edge-stable || true
-          sudo apt-get remove -y firefox || true
-          sudo apt-get remove -y powershell || true
-          sudo apt-get remove -y r-base-core || true
-          sudo apt-get autoremove -y
-          sudo apt-get clean
-          echo
-          echo "Listing top largest packages"
-          pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
-          head -n 30 <<< "${pkgs}"
-          echo
-          sudo rm -rfv build || true
-          df -h
+          sudo apt-get update \
+          && sudo apt-get install -y software-properties-common \
+          && sudo apt-get update \
+          && sudo add-apt-repository -y ppa:git-core/ppa \
+          && sudo apt-get update \
+          && sudo apt-get install -y git
      - name: Checkout
        uses: actions/checkout@v4
-
+      # - name: Release space from worker
+      #   run: |
+      #     echo "Listing top largest packages"
+      #     pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+      #     head -n 30 <<< "${pkgs}"
+      #     echo
+      #     df -h
+      #     echo
+      #     sudo apt-get remove -y '^llvm-.*|^libllvm.*' || true
+      #     sudo apt-get remove --auto-remove android-sdk-platform-tools || true
+      #     sudo apt-get purge --auto-remove android-sdk-platform-tools || true
+      #     sudo rm -rf /usr/local/lib/android
+      #     sudo apt-get remove -y '^dotnet-.*|^aspnetcore-.*' || true
+      #     sudo rm -rf /usr/share/dotnet
+      #     sudo apt-get remove -y '^mono-.*' || true
+      #     sudo apt-get remove -y '^ghc-.*' || true
+      #     sudo apt-get remove -y '.*jdk.*|.*jre.*' || true
+      #     sudo apt-get remove -y 'php.*' || true
+      #     sudo apt-get remove -y hhvm powershell firefox monodoc-manual msbuild || true
+      #     sudo apt-get remove -y '^google-.*' || true
+      #     sudo apt-get remove -y azure-cli || true
+      #     sudo apt-get remove -y '^mongo.*-.*|^postgresql-.*|^mysql-.*|^mssql-.*' || true
+      #     sudo apt-get remove -y '^gfortran-.*' || true
+      #     sudo apt-get remove -y microsoft-edge-stable || true
+      #     sudo apt-get remove -y firefox || true
+      #     sudo apt-get remove -y powershell || true
+      #     sudo apt-get remove -y r-base-core || true
+      #     sudo apt-get autoremove -y
+      #     sudo apt-get clean
+      #     echo
+      #     echo "Listing top largest packages"
+      #     pkgs=$(dpkg-query -Wf '${Installed-Size}\t${Package}\t${Status}\n' | awk '$NF == "installed"{print $1 "\t" $2}' | sort -nr)
+      #     head -n 30 <<< "${pkgs}"
+      #     echo
+      #     sudo rm -rfv build || true
+      #     df -h
      - name: Docker meta
        id: meta
        uses: docker/metadata-action@v5
--- a/.github/workflows/test-gpu.yml
+++ b/.github/workflows/test-gpu.yml
@@ -15,7 +15,7 @@ concurrency:

 jobs:
  ubuntu-latest:
-    runs-on: self-hosted
+    runs-on: gpu
    strategy:
      matrix:
        go-version: ['1.21.x']
@@ -59,5 +59,5 @@ jobs:
          make \
            TEST_DIR="/host/tests/${{ github.head_ref || github.ref }}" \
            teardown-e2e || true
-          sudo rm -rf /host/tests/${{ github.head_ref || github.ref }}
+          sudo rm -rf /host/tests/${{ github.head_ref || github.ref }} || true
          docker system prune -f -a --volumes || true
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@@ -14,7 +14,7 @@ concurrency:
  cancel-in-progress: true

 jobs:
-  ubuntu-latest:
+  tests-linux:
    runs-on: ubuntu-latest
    strategy:
      matrix:
@@ -67,41 +67,36 @@ jobs:
        run: |
          sudo apt-get update
          sudo apt-get install build-essential ffmpeg
-
+          curl https://repo.anaconda.com/pkgs/misc/gpgkeys/anaconda.asc | gpg --dearmor > conda.gpg && \
+             sudo install -o root -g root -m 644 conda.gpg /usr/share/keyrings/conda-archive-keyring.gpg && \
+              gpg --keyring /usr/share/keyrings/conda-archive-keyring.gpg --no-default-keyring --fingerprint 34161F5BF5EB1D4BFBBB8F0A8AEB4F8B29D82806 && \
+             sudo /bin/bash -c 'echo "deb [arch=amd64 signed-by=/usr/share/keyrings/conda-archive-keyring.gpg] https://repo.anaconda.com/pkgs/misc/debrepo/conda stable main" > /etc/apt/sources.list.d/conda.list' && \
+             sudo /bin/bash -c 'echo "deb [arch=amd64 signed-by=/usr/share/keyrings/conda-archive-keyring.gpg] https://repo.anaconda.com/pkgs/misc/debrepo/conda stable main" | tee -a /etc/apt/sources.list.d/conda.list' && \
+             sudo apt-get update && \
+             sudo apt-get install -y conda
          sudo apt-get install -y ca-certificates cmake curl patch
          sudo apt-get install -y libopencv-dev && sudo ln -s /usr/include/opencv4/opencv2 /usr/include/opencv2
-          sudo pip install -r extra/requirements.txt
+          
+          sudo rm -rfv /usr/bin/conda || true
+          PATH=$PATH:/opt/conda/bin make -C extra/grpc/huggingface

+          # Pre-build piper before we start tests in order to have shared libraries in place
+          make go-piper && \
+          GO_TAGS="tts" make -C go-piper piper.o && \
+          sudo cp -rfv go-piper/piper/build/pi/lib/. /usr/lib/ && \

-          # Pre-build stable diffusion before we install a newever version of abseil (not compatible with stablediffusion-ncn)
-          GO_TAGS="tts stablediffusion" GRPC_BACKENDS=backend-assets/grpc/stablediffusion make build
-
-          sudo mkdir /build && sudo chmod -R 777 /build && cd /build && \
-          curl -L "https://github.com/gabime/spdlog/archive/refs/tags/v1.11.0.tar.gz" | \
-          tar -xzvf - && \
-          mkdir -p "spdlog-1.11.0/build" && \
-          cd "spdlog-1.11.0/build" && \
-          cmake ..  && \
-          make -j8 && \
-          sudo cmake --install . --prefix /usr && mkdir -p "lib/Linux-$(uname -m)" && \
-          cd /build && \
-          mkdir -p "lib/Linux-$(uname -m)/piper_phonemize" && \
-          curl -L "https://github.com/rhasspy/piper-phonemize/releases/download/v1.0.0/libpiper_phonemize-amd64.tar.gz" | \
-          tar -C "lib/Linux-$(uname -m)/piper_phonemize" -xzvf - && ls -liah /build/lib/Linux-$(uname -m)/piper_phonemize/ && \
-          sudo cp -rfv /build/lib/Linux-$(uname -m)/piper_phonemize/lib/. /usr/lib/ && \
-          sudo ln -s /usr/lib/libpiper_phonemize.so /usr/lib/libpiper_phonemize.so.1 && \
-          sudo cp -rfv /build/lib/Linux-$(uname -m)/piper_phonemize/include/. /usr/include/
+          # Pre-build stable diffusion before we install a newer version of abseil (not compatible with stablediffusion-ncn)
+          GO_TAGS="stablediffusion tts" GRPC_BACKENDS=backend-assets/grpc/stablediffusion make build

          git clone --recurse-submodules -b v1.58.0 --depth 1 --shallow-submodules https://github.com/grpc/grpc && \
              cd grpc && mkdir -p cmake/build && cd cmake/build && cmake -DgRPC_INSTALL=ON \
                -DgRPC_BUILD_TESTS=OFF \
                ../.. && sudo make -j12 install
-
      - name: Test
        run: |
-          ESPEAK_DATA="/build/lib/Linux-$(uname -m)/piper_phonemize/lib/espeak-ng-data" GO_TAGS="tts stablediffusion" make test
+          GO_TAGS="stablediffusion tts" make test

-  macOS-latest:
+  tests-apple:
    runs-on: macOS-latest
    strategy:
      matrix:
--- a/.gitignore
+++ b/.gitignore
@@ -10,6 +10,8 @@ go-ggllm
 __pycache__/
 *.a
 get-sources
+/backend/cpp/llama/grpc-server
+/backend/cpp/llama/llama.cpp

 go-ggml-transformers
 go-gpt2
--- a/82
+++ b/82
@@ -8,13 +8,11 @@ FROM golang:$GO_VERSION as requirements-core
 ARG BUILD_TYPE
 ARG CUDA_MAJOR_VERSION=11
 ARG CUDA_MINOR_VERSION=7
-ARG SPDLOG_VERSION="1.11.0"
-ARG PIPER_PHONEMIZE_VERSION='1.0.0'
 ARG TARGETARCH
 ARG TARGETVARIANT

 ENV BUILD_TYPE=${BUILD_TYPE}
-ENV EXTERNAL_GRPC_BACKENDS="huggingface-embeddings:/build/extra/grpc/huggingface/huggingface.py,autogptq:/build/extra/grpc/autogptq/autogptq.py,bark:/build/extra/grpc/bark/ttsbark.py,diffusers:/build/extra/grpc/diffusers/backend_diffusers.py,exllama:/build/extra/grpc/exllama/exllama.py,vall-e-x:/build/extra/grpc/vall-e-x/ttsvalle.py,vllm:/build/extra/grpc/vllm/backend_vllm.py"
+ENV EXTERNAL_GRPC_BACKENDS="huggingface-embeddings:/build/extra/grpc/huggingface/run.sh,autogptq:/build/extra/grpc/autogptq/run.sh,bark:/build/extra/grpc/bark/run.sh,diffusers:/build/extra/grpc/diffusers/run.sh,exllama:/build/extra/grpc/exllama/run.sh,vall-e-x:/build/extra/grpc/vall-e-x/run.sh,vllm:/build/extra/grpc/vllm/run.sh"
 ENV GALLERIES='[{"name":"model-gallery", "url":"github:go-skynet/model-gallery/index.yaml"}, {"url": "github:go-skynet/model-gallery/huggingface.yaml","name":"huggingface"}]'
 ARG GO_TAGS="stablediffusion tts"

@@ -22,6 +20,9 @@ RUN apt-get update && \
    apt-get install -y ca-certificates curl patch pip cmake && apt-get clean


+COPY --chmod=644 custom-ca-certs/* /usr/local/share/ca-certificates/
+RUN update-ca-certificates
+
 # Use the variables in subsequent instructions
 RUN echo "Target Architecture: $TARGETARCH"
 RUN echo "Target Variant: $TARGETVARIANT"
@@ -49,42 +50,31 @@ RUN ln -s /usr/include/opencv4/opencv2 /usr/include/opencv2

 WORKDIR /build

-# piper requirements
-# Use pre-compiled Piper phonemization library (includes onnxruntime)
-#RUN if echo "${GO_TAGS}" | grep -q "tts"; then \
 RUN test -n "$TARGETARCH" \
    || (echo 'warn: missing $TARGETARCH, either set this `ARG` manually, or run using `docker buildkit`')

-RUN curl -L "https://github.com/gabime/spdlog/archive/refs/tags/v${SPDLOG_VERSION}.tar.gz" | \
-    tar -xzvf - && \
-    mkdir -p "spdlog-${SPDLOG_VERSION}/build" && \
-    cd "spdlog-${SPDLOG_VERSION}/build" && \
-    cmake ..  && \
-    make -j8 && \
-    cmake --install . --prefix /usr && mkdir -p "lib/Linux-$(uname -m)" && \
-    cd /build && \
-    mkdir -p "lib/Linux-$(uname -m)/piper_phonemize" && \
-    curl -L "https://github.com/rhasspy/piper-phonemize/releases/download/v${PIPER_PHONEMIZE_VERSION}/libpiper_phonemize-${TARGETARCH:-$(go env GOARCH)}${TARGETVARIANT}.tar.gz" | \
-    tar -C "lib/Linux-$(uname -m)/piper_phonemize" -xzvf - && ls -liah /build/lib/Linux-$(uname -m)/piper_phonemize/ && \
-    cp -rfv /build/lib/Linux-$(uname -m)/piper_phonemize/lib/. /usr/lib/ && \
-    ln -s /usr/lib/libpiper_phonemize.so /usr/lib/libpiper_phonemize.so.1 && \
-    cp -rfv /build/lib/Linux-$(uname -m)/piper_phonemize/include/. /usr/include/ && \
-    rm spdlog-${SPDLOG_VERSION} -rf
-
 # Extras requirements
 FROM requirements-core as requirements-extras

+RUN curl https://repo.anaconda.com/pkgs/misc/gpgkeys/anaconda.asc | gpg --dearmor > conda.gpg && \
+    install -o root -g root -m 644 conda.gpg /usr/share/keyrings/conda-archive-keyring.gpg && \
+    gpg --keyring /usr/share/keyrings/conda-archive-keyring.gpg --no-default-keyring --fingerprint 34161F5BF5EB1D4BFBBB8F0A8AEB4F8B29D82806 && \
+    echo "deb [arch=amd64 signed-by=/usr/share/keyrings/conda-archive-keyring.gpg] https://repo.anaconda.com/pkgs/misc/debrepo/conda stable main" > /etc/apt/sources.list.d/conda.list && \
+    echo "deb [arch=amd64 signed-by=/usr/share/keyrings/conda-archive-keyring.gpg] https://repo.anaconda.com/pkgs/misc/debrepo/conda stable main" | tee -a /etc/apt/sources.list.d/conda.list && \
+    apt-get update && \
+    apt-get install -y conda
+
 COPY extra/requirements.txt /build/extra/requirements.txt
 ENV PATH="/root/.cargo/bin:${PATH}"
 RUN pip install --upgrade pip
 RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
-RUN if [ "${TARGETARCH}" = "amd64" ]; then \
-        pip install git+https://github.com/suno-ai/bark.git diffusers invisible_watermark transformers accelerate safetensors;\
-    fi
-RUN if [ "${BUILD_TYPE}" = "cublas" ] && [ "${TARGETARCH}" = "amd64" ]; then \
-        pip install torch vllm && pip install auto-gptq https://github.com/jllllll/exllama/releases/download/0.0.10/exllama-0.0.10+cu${CUDA_MAJOR_VERSION}${CUDA_MINOR_VERSION}-cp39-cp39-linux_x86_64.whl;\
-    fi
-RUN pip install -r /build/extra/requirements.txt && rm -rf /build/extra/requirements.txt
+#RUN if [ "${TARGETARCH}" = "amd64" ]; then \
+#        pip install git+https://github.com/suno-ai/bark.git diffusers invisible_watermark transformers accelerate safetensors;\
+#    fi
+#RUN if [ "${BUILD_TYPE}" = "cublas" ] && [ "${TARGETARCH}" = "amd64" ]; then \
+#        pip install torch vllm && pip install auto-gptq https://github.com/jllllll/exllama/releases/download/0.0.10/exllama-0.0.10+cu${CUDA_MAJOR_VERSION}${CUDA_MINOR_VERSION}-cp39-cp39-linux_x86_64.whl;\
+ #   fi
+#RUN pip install -r /build/extra/requirements.txt && rm -rf /build/extra/requirements.txt

 # Vall-e-X
 RUN git clone https://github.com/Plachtaa/VALL-E-X.git /usr/lib/vall-e-x && cd /usr/lib/vall-e-x && pip install -r requirements.txt
@@ -126,7 +116,12 @@ RUN if [ "${BUILD_GRPC}" = "true" ]; then \
    ; fi

 # Rebuild with defaults backends
-RUN ESPEAK_DATA=/build/lib/Linux-$(uname -m)/piper_phonemize/lib/espeak-ng-data make build
+RUN make build
+
+RUN if [ ! -d "/build/go-piper/piper/build/pi/lib/" ]; then \
+    mkdir -p /build/go-piper/piper/build/pi/lib/ \
+    touch /build/go-piper/piper/build/pi/lib/keep \
+    ; fi

 ###################################
 ###################################
@@ -136,6 +131,7 @@ FROM requirements-${IMAGE_TYPE}
 ARG FFMPEG
 ARG BUILD_TYPE
 ARG TARGETARCH
+ARG IMAGE_TYPE=extras

 ENV BUILD_TYPE=${BUILD_TYPE}
 ENV REBUILD=false
@@ -163,9 +159,35 @@ RUN make prepare-sources
 # Copy the binary
 COPY --from=builder /build/local-ai ./

+# Copy shared libraries for piper
+COPY --from=builder /build/go-piper/piper/build/pi/lib/* /usr/lib/
+
 # do not let stablediffusion rebuild (requires an older version of absl)
 COPY --from=builder /build/backend-assets/grpc/stablediffusion ./backend-assets/grpc/stablediffusion

+## Duplicated from Makefile to avoid having a big layer that's hard to push
+RUN if [ "${IMAGE_TYPE}" = "extras" ]; then \
+	PATH=$PATH:/opt/conda/bin make -C extra/grpc/autogptq \
+    ; fi
+RUN if [ "${IMAGE_TYPE}" = "extras" ]; then \
+	PATH=$PATH:/opt/conda/bin make -C extra/grpc/bark \
+    ; fi
+RUN if [ "${IMAGE_TYPE}" = "extras" ]; then \
+	PATH=$PATH:/opt/conda/bin make -C extra/grpc/diffusers \
+    ; fi
+RUN if [ "${IMAGE_TYPE}" = "extras" ]; then \
+	PATH=$PATH:/opt/conda/bin make -C extra/grpc/vllm \
+    ; fi
+RUN if [ "${IMAGE_TYPE}" = "extras" ]; then \
+	PATH=$PATH:/opt/conda/bin make -C extra/grpc/huggingface \
+    ; fi
+RUN if [ "${IMAGE_TYPE}" = "extras" ]; then \
+	PATH=$PATH:/opt/conda/bin make -C extra/grpc/vall-e-x \
+    ; fi
+RUN if [ "${IMAGE_TYPE}" = "extras" ]; then \
+	PATH=$PATH:/opt/conda/bin make -C extra/grpc/exllama \
+    ; fi
+
 # Copy VALLE-X as it's not a real "lib"
 RUN if [ -d /usr/lib/vall-e-x ]; then \
    cp -rfv /usr/lib/vall-e-x/* ./ ; \ 
--- a/73
+++ b/73
@@ -8,7 +8,7 @@ GOLLAMA_VERSION?=aeba71ee842819da681ea537e78846dc75949ac0

 GOLLAMA_STABLE_VERSION?=50cee7712066d9e38306eccadcfbb44ea87df4b7

-CPPLLAMA_VERSION?=9d02956443e5c1ded29b7b5ed8a21bc01ba6f563
+CPPLLAMA_VERSION?=a75fa576abba9d37f463580c379e4bbf1e1ad03c

 # gpt4all version
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
@@ -28,7 +28,7 @@ WHISPER_CPP_VERSION?=85ed71aaec8e0612a84c0b67804bde75aa75a273
 BERT_VERSION?=6abe312cded14042f6b7c3cd8edf082713334a4d

 # go-piper version
-PIPER_VERSION?=56b8a81b4760a6fbee1a82e62f007ae7e8f010a7
+PIPER_VERSION?=736f6fb639ab8e3397356e48eeb6bdcb9da88a78

 # stablediffusion version
 STABLEDIFFUSION_VERSION?=d89260f598afb809279bc72aa0107b4292587632
@@ -52,7 +52,6 @@ override LD_FLAGS += -X "github.com/go-skynet/LocalAI/internal.Version=$(VERSION
 override LD_FLAGS += -X "github.com/go-skynet/LocalAI/internal.Commit=$(shell git rev-parse HEAD)"

 OPTIONAL_TARGETS?=
-ESPEAK_DATA?=

 OS := $(shell uname -s)
 ARCH := $(shell uname -m)
@@ -120,6 +119,8 @@ endif
 ifeq ($(findstring tts,$(GO_TAGS)),tts)
 #	OPTIONAL_TARGETS+=go-piper/libpiper_binding.a
 #	OPTIONAL_TARGETS+=backend-assets/espeak-ng-data
+	PIPER_CGO_CXXFLAGS+=-I$(shell pwd)/go-piper/piper/src/cpp -I$(shell pwd)/go-piper/piper/build/fi/include -I$(shell pwd)/go-piper/piper/build/pi/include -I$(shell pwd)/go-piper/piper/build/si/include
+ 	PIPER_CGO_LDFLAGS+=-L$(shell pwd)/go-piper/piper/build/fi/lib -L$(shell pwd)/go-piper/piper/build/pi/lib -L$(shell pwd)/go-piper/piper/build/si/lib -lfmt -lspdlog
 	OPTIONAL_GRPC+=backend-assets/grpc/piper
 endif

@@ -175,14 +176,10 @@ backend-assets/gpt4all: gpt4all/gpt4all-bindings/golang/libgpt4all.a
 	@cp gpt4all/gpt4all-bindings/golang/buildllm/*.dylib backend-assets/gpt4all/ || true
 	@cp gpt4all/gpt4all-bindings/golang/buildllm/*.dll backend-assets/gpt4all/ || true

-backend-assets/espeak-ng-data:
+backend-assets/espeak-ng-data: go-piper
 	mkdir -p backend-assets/espeak-ng-data
-ifdef ESPEAK_DATA
-	@cp -rf $(ESPEAK_DATA)/. backend-assets/espeak-ng-data
-else
-	@echo "ESPEAK_DATA not set, skipping tts. Note that this will break the tts functionality."
-	@touch backend-assets/espeak-ng-data/keep
-endif
+	$(MAKE) -C go-piper piper.o
+	@cp -rf go-piper/piper/build/pi/share/espeak-ng-data/. backend-assets/espeak-ng-data

 gpt4all/gpt4all-bindings/golang/libgpt4all.a: gpt4all
 	$(MAKE) -C gpt4all/gpt4all-bindings/golang/ libgpt4all.a
@@ -267,6 +264,9 @@ clean: ## Remove build related file
 	rm -rf ./go-piper
 	rm -rf $(BINARY_NAME)
 	rm -rf release/
+	rm -rf ./backend/cpp/grpc/grpc_repo
+	rm -rf ./backend/cpp/grpc/build
+	rm -rf ./backend/cpp/grpc/installed_packages
 	$(MAKE) -C backend/cpp/llama clean

 ## Build:
@@ -290,12 +290,12 @@ run: prepare ## run local-ai
 test-models/testmodel:
 	mkdir test-models
 	mkdir test-dir
-	wget https://huggingface.co/nnakasato/ggml-model-test/resolve/main/ggml-model-q4.bin -O test-models/testmodel
-	wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin -O test-models/whisper-en
-	wget https://huggingface.co/skeskinen/ggml/resolve/main/all-MiniLM-L6-v2/ggml-model-q4_0.bin -O test-models/bert
-	wget https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav -O test-dir/audio.wav
-	wget https://huggingface.co/mudler/rwkv-4-raven-1.5B-ggml/resolve/main/RWKV-4-Raven-1B5-v11-Eng99%2525-Other1%2525-20230425-ctx4096_Q4_0.bin -O test-models/rwkv
-	wget https://raw.githubusercontent.com/saharNooby/rwkv.cpp/5eb8f09c146ea8124633ab041d9ea0b1f1db4459/rwkv/20B_tokenizer.json -O test-models/rwkv.tokenizer.json
+	wget -q https://huggingface.co/nnakasato/ggml-model-test/resolve/main/ggml-model-q4.bin -O test-models/testmodel
+	wget -q https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin -O test-models/whisper-en
+	wget -q https://huggingface.co/mudler/all-MiniLM-L6-v2/resolve/main/ggml-model-q4_0.bin -O test-models/bert
+	wget -q https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav -O test-dir/audio.wav
+	wget -q https://huggingface.co/mudler/rwkv-4-raven-1.5B-ggml/resolve/main/RWKV-4-Raven-1B5-v11-Eng99%2525-Other1%2525-20230425-ctx4096_Q4_0.bin -O test-models/rwkv
+	wget -q https://raw.githubusercontent.com/saharNooby/rwkv.cpp/5eb8f09c146ea8124633ab041d9ea0b1f1db4459/rwkv/20B_tokenizer.json -O test-models/rwkv.tokenizer.json
 	cp tests/models_fixtures/* test-models

 prepare-test: grpcs
@@ -306,8 +306,8 @@ test: prepare test-models/testmodel grpcs
 	@echo 'Running tests'
 	export GO_TAGS="tts stablediffusion"
 	$(MAKE) prepare-test
-	HUGGINGFACE_GRPC=$(abspath ./)/extra/grpc/huggingface/huggingface.py TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
-	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="!gpt4all && !llama && !llama-gguf" --flake-attempts 5 -v -r ./api ./pkg
+	HUGGINGFACE_GRPC=$(abspath ./)/extra/grpc/huggingface/run.sh TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
+	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="!gpt4all && !llama && !llama-gguf"  --flake-attempts 5 --fail-fast -v -r ./api ./pkg
 	$(MAKE) test-gpt4all
 	$(MAKE) test-llama
 	$(MAKE) test-llama-gguf
@@ -386,6 +386,16 @@ protogen-python:
 	python3 -m grpc_tools.protoc -Ipkg/grpc/proto/ --python_out=extra/grpc/vllm/ --grpc_python_out=extra/grpc/vllm/ pkg/grpc/proto/backend.proto

 ## GRPC
+# Note: it is duplicated in the Dockerfile
+prepare-extra-conda-environments:
+	$(MAKE) -C extra/grpc/autogptq
+	$(MAKE) -C extra/grpc/bark
+	$(MAKE) -C extra/grpc/diffusers
+	$(MAKE) -C extra/grpc/vllm
+	$(MAKE) -C extra/grpc/huggingface
+	$(MAKE) -C extra/grpc/vall-e-x
+	$(MAKE) -C extra/grpc/exllama
+

 backend-assets/grpc:
 	mkdir -p backend-assets/grpc
@@ -399,9 +409,30 @@ ifeq ($(BUILD_TYPE),metal)
 	cp go-llama/build/bin/ggml-metal.metal backend-assets/grpc/
 endif

-backend/cpp/llama/grpc-server:
-	LLAMA_VERSION=$(CPPLLAMA_VERSION) $(MAKE) -C backend/cpp/llama grpc-server
+## BACKEND CPP LLAMA START
+# Sets the variables in case it has to build the gRPC locally.
+INSTALLED_PACKAGES=$(CURDIR)/backend/cpp/grpc/installed_packages
+INSTALLED_LIB_CMAKE=$(INSTALLED_PACKAGES)/lib/cmake
+ADDED_CMAKE_ARGS=-Dabsl_DIR=${INSTALLED_LIB_CMAKE}/absl \
+                 -DProtobuf_DIR=${INSTALLED_LIB_CMAKE}/protobuf \
+                 -Dutf8_range_DIR=${INSTALLED_LIB_CMAKE}/utf8_range \
+                 -DgRPC_DIR=${INSTALLED_LIB_CMAKE}/grpc \
+                 -DCMAKE_CXX_STANDARD_INCLUDE_DIRECTORIES=${INSTALLED_PACKAGES}/include

+backend/cpp/llama/grpc-server:
+ifdef BUILD_GRPC_FOR_BACKEND_LLAMA
+	backend/cpp/grpc/script/build_grpc.sh ${INSTALLED_PACKAGES}
+	export _PROTOBUF_PROTOC=${INSTALLED_PACKAGES}/bin/proto && \
+	export _GRPC_CPP_PLUGIN_EXECUTABLE=${INSTALLED_PACKAGES}/bin/grpc_cpp_plugin && \
+	export PATH=${PATH}:${INSTALLED_PACKAGES}/bin && \
+	CMAKE_ARGS="${ADDED_CMAKE_ARGS}" LLAMA_VERSION=$(CPPLLAMA_VERSION) $(MAKE) -C backend/cpp/llama grpc-server 
+else
+	echo "BUILD_GRPC_FOR_BACKEND_LLAMA is not defined."
+	LLAMA_VERSION=$(CPPLLAMA_VERSION) $(MAKE) -C backend/cpp/llama grpc-server			
+endif
+## BACKEND CPP LLAMA END
+		
+##
 backend-assets/grpc/llama-cpp: backend-assets/grpc backend/cpp/llama/grpc-server
 	cp -rfv backend/cpp/llama/grpc-server backend-assets/grpc/llama-cpp
 # TODO: every binary should have its own folder instead, so can have different metal implementations
@@ -469,7 +500,7 @@ backend-assets/grpc/stablediffusion: backend-assets/grpc
 	fi

 backend-assets/grpc/piper: backend-assets/grpc backend-assets/espeak-ng-data go-piper/libpiper_binding.a
-	CGO_LDFLAGS="$(CGO_LDFLAGS)" LIBRARY_PATH=$(shell pwd)/go-piper \
+	CGO_CXXFLAGS="$(PIPER_CGO_CXXFLAGS)" CGO_LDFLAGS="$(PIPER_CGO_LDFLAGS)" LIBRARY_PATH=$(shell pwd)/go-piper \
 	$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/piper ./cmd/grpc/piper/

 backend-assets/grpc/whisper: backend-assets/grpc whisper.cpp/libwhisper.a
--- a/api/api_test.go
+++ b/api/api_test.go
@@ -457,7 +457,7 @@ var _ = Describe("API test", func() {
 				Eventually(func() bool {
 					response := getModelStatus("http://127.0.0.1:9090/models/jobs/" + uuid)
 					return response["processed"].(bool)
-				}, "360s", "10s").Should(Equal(true))
+				}, "960s", "10s").Should(Equal(true))

 				resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "gpt4all-j", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "How are you?"}}})
 				Expect(err).ToNot(HaveOccurred())
@@ -687,7 +687,7 @@ var _ = Describe("API test", func() {
 					Input: []string{"sun", "cat"},
 				},
 			)
-			Expect(err).ToNot(HaveOccurred())
+			Expect(err).ToNot(HaveOccurred(), err)
 			Expect(len(resp.Data[0].Embedding)).To(BeNumerically("==", 384))
 			Expect(len(resp.Data[1].Embedding)).To(BeNumerically("==", 384))

--- a/api/backend/image.go
+++ b/api/backend/image.go
@@ -21,6 +21,7 @@ func ImageGeneration(height, width, mode, step, seed int, positive_prompt, negat
 			PipelineType:  c.Diffusers.PipelineType,
 			CFGScale:      c.Diffusers.CFGScale,
 			LoraAdapter:   c.LoraAdapter,
+			LoraScale:     c.LoraScale,
 			LoraBase:      c.LoraBase,
 			IMG2IMG:       c.Diffusers.IMG2IMG,
 			CLIPModel:     c.Diffusers.ClipModel,
--- a/api/backend/llm.go
+++ b/api/backend/llm.go
@@ -26,7 +26,7 @@ type TokenUsage struct {
 	Completion int
 }

-func ModelInference(ctx context.Context, s string, loader *model.ModelLoader, c config.Config, o *options.Option, tokenCallback func(string, TokenUsage) bool) (func() (LLMResponse, error), error) {
+func ModelInference(ctx context.Context, s string, images []string, loader *model.ModelLoader, c config.Config, o *options.Option, tokenCallback func(string, TokenUsage) bool) (func() (LLMResponse, error), error) {
 	modelFile := c.Model

 	grpcOpts := gRPCModelOpts(c)
@@ -72,6 +72,7 @@ func ModelInference(ctx context.Context, s string, loader *model.ModelLoader, c
 	fn := func() (LLMResponse, error) {
 		opts := gRPCPredictOpts(c, loader.ModelPath)
 		opts.Prompt = s
+		opts.Images = images

 		tokenUsage := TokenUsage{}

--- a/api/backend/options.go
+++ b/api/backend/options.go
@@ -38,29 +38,35 @@ func gRPCModelOpts(c config.Config) *pb.ModelOptions {
 	}

 	return &pb.ModelOptions{
-		ContextSize:   int32(c.ContextSize),
-		Seed:          int32(c.Seed),
-		NBatch:        int32(b),
-		NoMulMatQ:     c.NoMulMatQ,
-		DraftModel:    c.DraftModel,
-		AudioPath:     c.VallE.AudioPath,
-		Quantization:  c.Quantization,
-		LoraAdapter:   c.LoraAdapter,
-		LoraBase:      c.LoraBase,
-		NGQA:          c.NGQA,
-		RMSNormEps:    c.RMSNormEps,
-		F16Memory:     c.F16,
-		MLock:         c.MMlock,
-		RopeFreqBase:  c.RopeFreqBase,
-		RopeFreqScale: c.RopeFreqScale,
-		NUMA:          c.NUMA,
-		Embeddings:    c.Embeddings,
-		LowVRAM:       c.LowVRAM,
-		NGPULayers:    int32(c.NGPULayers),
-		MMap:          c.MMap,
-		MainGPU:       c.MainGPU,
-		Threads:       int32(c.Threads),
-		TensorSplit:   c.TensorSplit,
+		ContextSize:    int32(c.ContextSize),
+		Seed:           int32(c.Seed),
+		NBatch:         int32(b),
+		NoMulMatQ:      c.NoMulMatQ,
+		DraftModel:     c.DraftModel,
+		AudioPath:      c.VallE.AudioPath,
+		Quantization:   c.Quantization,
+		MMProj:         c.MMProj,
+		YarnExtFactor:  c.YarnExtFactor,
+		YarnAttnFactor: c.YarnAttnFactor,
+		YarnBetaFast:   c.YarnBetaFast,
+		YarnBetaSlow:   c.YarnBetaSlow,
+		LoraAdapter:    c.LoraAdapter,
+		LoraBase:       c.LoraBase,
+		LoraScale:      c.LoraScale,
+		NGQA:           c.NGQA,
+		RMSNormEps:     c.RMSNormEps,
+		F16Memory:      c.F16,
+		MLock:          c.MMlock,
+		RopeFreqBase:   c.RopeFreqBase,
+		RopeFreqScale:  c.RopeFreqScale,
+		NUMA:           c.NUMA,
+		Embeddings:     c.Embeddings,
+		LowVRAM:        c.LowVRAM,
+		NGPULayers:     int32(c.NGPULayers),
+		MMap:           c.MMap,
+		MainGPU:        c.MainGPU,
+		Threads:        int32(c.Threads),
+		TensorSplit:    c.TensorSplit,
 		// AutoGPTQ
 		ModelBaseName:    c.AutoGPTQ.ModelBaseName,
 		Device:           c.AutoGPTQ.Device,
--- a/api/config/config.go
+++ b/api/config/config.go
@@ -100,10 +100,18 @@ type LLMConfig struct {
 	NUMA            bool     `yaml:"numa"`
 	LoraAdapter     string   `yaml:"lora_adapter"`
 	LoraBase        string   `yaml:"lora_base"`
+	LoraScale       float32  `yaml:"lora_scale"`
 	NoMulMatQ       bool     `yaml:"no_mulmatq"`
 	DraftModel      string   `yaml:"draft_model"`
 	NDraft          int32    `yaml:"n_draft"`
 	Quantization    string   `yaml:"quantization"`
+	MMProj          string   `yaml:"mmproj"`
+
+	RopeScaling    string  `yaml:"rope_scaling"`
+	YarnExtFactor  float32 `yaml:"yarn_ext_factor"`
+	YarnAttnFactor float32 `yaml:"yarn_attn_factor"`
+	YarnBetaFast   float32 `yaml:"yarn_beta_fast"`
+	YarnBetaSlow   float32 `yaml:"yarn_beta_slow"`
 }

 type AutoGPTQ struct {
--- a/api/openai/chat.go
+++ b/api/openai/chat.go
@@ -81,6 +81,10 @@ func ChatEndpoint(cm *config.ConfigLoader, o *options.Option) func(c *fiber.Ctx)
 			noActionDescription = config.FunctionsConfig.NoActionDescriptionName
 		}

+		if input.ResponseFormat == "json_object" {
+			input.Grammar = grammar.JSONBNF
+		}
+
 		// process functions if we have any defined or if we have a function call string
 		if len(input.Functions) > 0 && config.ShouldUseFunctions() {
 			log.Debug().Msgf("Response needs to process functions")
@@ -140,14 +144,14 @@ func ChatEndpoint(cm *config.ConfigLoader, o *options.Option) func(c *fiber.Ctx)
 				}
 			}
 			r := config.Roles[role]
-			contentExists := i.Content != nil && *i.Content != ""
+			contentExists := i.Content != nil && i.StringContent != ""
 			// First attempt to populate content via a chat message specific template
 			if config.TemplateConfig.ChatMessage != "" {
 				chatMessageData := model.ChatMessageTemplateData{
 					SystemPrompt: config.SystemPrompt,
 					Role:         r,
 					RoleName:     role,
-					Content:      *i.Content,
+					Content:      i.StringContent,
 					MessageIndex: messageIndex,
 				}
 				templatedChatMessage, err := o.Loader.EvaluateTemplateForChatMessage(config.TemplateConfig.ChatMessage, chatMessageData)
@@ -166,7 +170,7 @@ func ChatEndpoint(cm *config.ConfigLoader, o *options.Option) func(c *fiber.Ctx)
 			if content == "" {
 				if r != "" {
 					if contentExists {
-						content = fmt.Sprint(r, " ", *i.Content)
+						content = fmt.Sprint(r, i.StringContent)
 					}
 					if i.FunctionCall != nil {
 						j, err := json.Marshal(i.FunctionCall)
@@ -180,7 +184,7 @@ func ChatEndpoint(cm *config.ConfigLoader, o *options.Option) func(c *fiber.Ctx)
 					}
 				} else {
 					if contentExists {
-						content = fmt.Sprint(*i.Content)
+						content = fmt.Sprint(i.StringContent)
 					}
 					if i.FunctionCall != nil {
 						j, err := json.Marshal(i.FunctionCall)
@@ -334,7 +338,11 @@ func ChatEndpoint(cm *config.ConfigLoader, o *options.Option) func(c *fiber.Ctx)
 					// Otherwise ask the LLM to understand the JSON output and the context, and return a message
 					// Note: This costs (in term of CPU) another computation
 					config.Grammar = ""
-					predFunc, err := backend.ModelInference(input.Context, predInput, o.Loader, *config, o, nil)
+					images := []string{}
+					for _, m := range input.Messages {
+						images = append(images, m.StringImages...)
+					}
+					predFunc, err := backend.ModelInference(input.Context, predInput, images, o.Loader, *config, o, nil)
 					if err != nil {
 						log.Error().Msgf("inference error: %s", err.Error())
 						return
--- a/api/openai/completion.go
+++ b/api/openai/completion.go
@@ -12,6 +12,7 @@ import (
 	config "github.com/go-skynet/LocalAI/api/config"
 	"github.com/go-skynet/LocalAI/api/options"
 	"github.com/go-skynet/LocalAI/api/schema"
+	"github.com/go-skynet/LocalAI/pkg/grammar"
 	model "github.com/go-skynet/LocalAI/pkg/model"
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
@@ -64,6 +65,10 @@ func CompletionEndpoint(cm *config.ConfigLoader, o *options.Option) func(c *fibe
 			return fmt.Errorf("failed reading parameters from request:%w", err)
 		}

+		if input.ResponseFormat == "json_object" {
+			input.Grammar = grammar.JSONBNF
+		}
+
 		log.Debug().Msgf("Parameter Config: %+v", config)

 		if input.Stream {
--- a/api/openai/inference.go
+++ b/api/openai/inference.go
@@ -23,8 +23,13 @@ func ComputeChoices(
 		n = 1
 	}

+	images := []string{}
+	for _, m := range req.Messages {
+		images = append(images, m.StringImages...)
+	}
+
 	// get the model function to call for the result
-	predFunc, err := backend.ModelInference(req.Context, predInput, loader, *config, o, tokenCallback)
+	predFunc, err := backend.ModelInference(req.Context, predInput, images, loader, *config, o, tokenCallback)
 	if err != nil {
 		return result, backend.TokenUsage{}, err
 	}
--- a/api/openai/request.go
+++ b/api/openai/request.go
@@ -2,8 +2,11 @@ package openai

 import (
 	"context"
+	"encoding/base64"
 	"encoding/json"
 	"fmt"
+	"io/ioutil"
+	"net/http"
 	"os"
 	"path/filepath"
 	"strings"
@@ -24,7 +27,7 @@ func readInput(c *fiber.Ctx, o *options.Option, randomModel bool) (string, *sche
 	input.Cancel = cancel
 	// Get input data from the request body
 	if err := c.BodyParser(input); err != nil {
-		return "", nil, err
+		return "", nil, fmt.Errorf("failed parsing request body: %w", err)
 	}

 	modelFile := input.Model
@@ -61,6 +64,37 @@ func readInput(c *fiber.Ctx, o *options.Option, randomModel bool) (string, *sche
 	return modelFile, input, nil
 }

+// this function check if the string is an URL, if it's an URL downloads the image in memory
+// encodes it in base64 and returns the base64 string
+func getBase64Image(s string) (string, error) {
+	if strings.HasPrefix(s, "http") {
+		// download the image
+		resp, err := http.Get(s)
+		if err != nil {
+			return "", err
+		}
+		defer resp.Body.Close()
+
+		// read the image data into memory
+		data, err := ioutil.ReadAll(resp.Body)
+		if err != nil {
+			return "", err
+		}
+
+		// encode the image data in base64
+		encoded := base64.StdEncoding.EncodeToString(data)
+
+		// return the base64 string
+		return encoded, nil
+	}
+
+	// if the string instead is prefixed with "data:image/jpeg;base64,", drop it
+	if strings.HasPrefix(s, "data:image/jpeg;base64,") {
+		return strings.ReplaceAll(s, "data:image/jpeg;base64,", ""), nil
+	}
+	return "", fmt.Errorf("not valid string")
+}
+
 func updateConfig(config *config.Config, input *schema.OpenAIRequest) {
 	if input.Echo {
 		config.Echo = input.Echo
@@ -129,6 +163,35 @@ func updateConfig(config *config.Config, input *schema.OpenAIRequest) {
 		}
 	}

+	// Decode each request's message content
+	index := 0
+	for i, m := range input.Messages {
+		switch content := m.Content.(type) {
+		case string:
+			input.Messages[i].StringContent = content
+		case []interface{}:
+			dat, _ := json.Marshal(content)
+			c := []schema.Content{}
+			json.Unmarshal(dat, &c)
+			for _, pp := range c {
+				if pp.Type == "text" {
+					input.Messages[i].StringContent = pp.Text
+				} else if pp.Type == "image_url" {
+					// Detect if pp.ImageURL is an URL, if it is download the image and encode it in base64:
+					base64, err := getBase64Image(pp.ImageURL.URL)
+					if err == nil {
+						input.Messages[i].StringImages = append(input.Messages[i].StringImages, base64) // TODO: make sure that we only return base64 stuff
+						// set a placeholder for each image
+						input.Messages[i].StringContent = fmt.Sprintf("[img-%d]", index) + input.Messages[i].StringContent
+						index++
+					} else {
+						fmt.Print("Failed encoding image", err)
+					}
+				}
+			}
+		}
+	}
+
 	if input.RepeatPenalty != 0 {
 		config.RepeatPenalty = input.RepeatPenalty
 	}
--- a/api/schema/openai.go
+++ b/api/schema/openai.go
@@ -55,11 +55,25 @@ type Choice struct {
 	Text         string   `json:"text,omitempty"`
 }

+type Content struct {
+	Type     string     `json:"type" yaml:"type"`
+	Text     string     `json:"text" yaml:"text"`
+	ImageURL ContentURL `json:"image_url" yaml:"image_url"`
+}
+
+type ContentURL struct {
+	URL string `json:"url" yaml:"url"`
+}
+
 type Message struct {
 	// The message role
 	Role string `json:"role,omitempty" yaml:"role"`
 	// The message content
-	Content *string `json:"content" yaml:"content"`
+	Content interface{} `json:"content" yaml:"content"`
+
+	StringContent string   `json:"string_content,omitempty" yaml:"string_content,omitempty"`
+	StringImages  []string `json:"string_images,omitempty" yaml:"string_images,omitempty"`
+
 	// A result of a function call
 	FunctionCall interface{} `json:"function_call,omitempty" yaml:"function_call,omitempty"`
 }
--- a/backend/cpp/grpc/.gitignore
+++ b/backend/cpp/grpc/.gitignore
@@ -0,0 +1,3 @@
+installed_packages/
+grpc_build/
+grpc_repo/
--- a/backend/cpp/grpc/script/build_grpc.sh
+++ b/backend/cpp/grpc/script/build_grpc.sh
@@ -0,0 +1,81 @@
+#!/bin/bash
+
+# Builds locally from sources the packages needed by the llama cpp backend.
+
+# Makes sure a few base packages exist.
+# sudo apt-get --no-upgrade -y install g++ gcc binutils cmake git build-essential autoconf libtool pkg-config 
+
+SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )"
+echo "Script directory: $SCRIPT_DIR"
+
+CPP_INSTALLED_PACKAGES_DIR=$1
+if [ -z ${CPP_INSTALLED_PACKAGES_DIR} ]; then 
+    echo "CPP_INSTALLED_PACKAGES_DIR env variable not set. Don't know where to install: failed."; 
+    echo
+    exit -1
+fi
+
+if [ -d "${CPP_INSTALLED_PACKAGES_DIR}" ]; then
+  echo "gRPC installation directory already exists. Nothing to do."
+  exit 0
+fi
+
+# The depth when cloning a git repo. 1 speeds up the clone when the repo history is not needed.
+GIT_CLONE_DEPTH=1
+
+NUM_BUILD_THREADS=$(nproc --ignore=1)
+
+# Google gRPC --------------------------------------------------------------------------------------
+TAG_LIB_GRPC="v1.59.0"
+GIT_REPO_LIB_GRPC="https://github.com/grpc/grpc.git"
+GRPC_REPO_DIR="${SCRIPT_DIR}/../grpc_repo"
+GRPC_BUILD_DIR="${SCRIPT_DIR}/../grpc_build"
+SRC_DIR_LIB_GRPC="${GRPC_REPO_DIR}/grpc"
+
+echo "SRC_DIR_LIB_GRPC: ${SRC_DIR_LIB_GRPC}"
+echo "GRPC_REPO_DIR: ${GRPC_REPO_DIR}"
+echo "GRPC_BUILD_DIR: ${GRPC_BUILD_DIR}"
+
+mkdir -pv ${GRPC_REPO_DIR}
+
+rm   -rf ${GRPC_BUILD_DIR}
+mkdir -pv ${GRPC_BUILD_DIR}
+
+mkdir -pv ${CPP_INSTALLED_PACKAGES_DIR}
+	
+if [ -d "${SRC_DIR_LIB_GRPC}" ]; then
+  echo "gRPC source already exists locally. Not cloned again."
+else  
+  ( cd ${GRPC_REPO_DIR} && \
+    git clone --depth ${GIT_CLONE_DEPTH} -b ${TAG_LIB_GRPC} ${GIT_REPO_LIB_GRPC} && \
+    cd ${SRC_DIR_LIB_GRPC} && \
+    git submodule update --init --recursive --depth ${GIT_CLONE_DEPTH} 
+  )    
+fi
+
+( cd ${GRPC_BUILD_DIR} && \
+  cmake -G "Unix Makefiles" \
+     -DCMAKE_BUILD_TYPE=Release \
+     -DgRPC_INSTALL=ON \
+     -DEXECUTABLE_OUTPUT_PATH=${CPP_INSTALLED_PACKAGES_DIR}/grpc/bin \
+     -DLIBRARY_OUTPUT_PATH=${CPP_INSTALLED_PACKAGES_DIR}/grpc/lib \
+     -DgRPC_BUILD_TESTS=OFF \
+     -DgRPC_BUILD_CSHARP_EXT=OFF \
+     -DgRPC_BUILD_GRPC_CPP_PLUGIN=ON \
+     -DgRPC_BUILD_GRPC_CSHARP_PLUGIN=OFF \
+     -DgRPC_BUILD_GRPC_NODE_PLUGIN=OFF \
+     -DgRPC_BUILD_GRPC_OBJECTIVE_C_PLUGIN=OFF \
+     -DgRPC_BUILD_GRPC_PHP_PLUGIN=OFF \
+     -DgRPC_BUILD_GRPC_PYTHON_PLUGIN=ON \
+     -DgRPC_BUILD_GRPC_RUBY_PLUGIN=OFF \
+     -Dprotobuf_WITH_ZLIB=ON \
+     -DRE2_BUILD_TESTING=OFF \
+     -DCMAKE_INSTALL_PREFIX=${CPP_INSTALLED_PACKAGES_DIR}/ \
+     ${SRC_DIR_LIB_GRPC}  && \
+  cmake --build .  -- -j ${NUM_BUILD_THREADS} && \
+  cmake --build .  --target install -- -j ${NUM_BUILD_THREADS} 
+)
+
+rm -rf ${GRPC_BUILD_DIR}
+rm -rf ${GRPC_REPO_DIR}
+
--- a/backend/cpp/llama/CMakeLists.txt
+++ b/backend/cpp/llama/CMakeLists.txt
@@ -1,9 +1,22 @@
+
+## XXX: In some versions of CMake clip wasn't being built before llama.
+## This is an hack for now, but it should be fixed in the future.
+set(TARGET myclip)
+add_library(${TARGET} clip.cpp clip.h)
+install(TARGETS ${TARGET} LIBRARY)
+target_link_libraries(${TARGET} PRIVATE common ggml ${CMAKE_THREAD_LIBS_INIT})
+target_compile_features(${TARGET} PRIVATE cxx_std_11)
+if (NOT MSVC)
+    target_compile_options(${TARGET} PRIVATE -Wno-cast-qual) # stb_image.h
+endif()
+
+set(TARGET grpc-server)
+# END CLIP hack
 set(CMAKE_CXX_STANDARD 17)
 cmake_minimum_required(VERSION 3.15)
 set(TARGET grpc-server)
 set(_PROTOBUF_LIBPROTOBUF libprotobuf)
 set(_REFLECTION grpc++_reflection)
-
 if (${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
    link_directories("/opt/homebrew/lib")
    include_directories("/opt/homebrew/include")
@@ -47,10 +60,10 @@ add_library(hw_grpc_proto
  ${hw_grpc_srcs}
  ${hw_grpc_hdrs}
  ${hw_proto_srcs}
-  ${hw_proto_hdrs})
+  ${hw_proto_hdrs} )

-add_executable(${TARGET} grpc-server.cpp)
-target_link_libraries(${TARGET} PRIVATE common llama ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
+add_executable(${TARGET} grpc-server.cpp json.hpp )
+target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
  absl::flags_parse
  gRPC::${_REFLECTION}
  gRPC::${_GRPC_GRPCPP}
--- a/backend/cpp/llama/Makefile
+++ b/backend/cpp/llama/Makefile
@@ -1,5 +1,5 @@

-LLAMA_VERSION?=
+LLAMA_VERSION?=d9b33fe95bd257b36c84ee5769cc048230067d6f

 CMAKE_ARGS?=
 BUILD_TYPE?=
@@ -27,11 +27,17 @@ llama.cpp/examples/grpc-server:
 	mkdir -p llama.cpp/examples/grpc-server
 	cp -r $(abspath ./)/CMakeLists.txt llama.cpp/examples/grpc-server/
 	cp -r $(abspath ./)/grpc-server.cpp llama.cpp/examples/grpc-server/
+	cp -rfv $(abspath ./)/json.hpp llama.cpp/examples/grpc-server/
 	echo "add_subdirectory(grpc-server)" >> llama.cpp/examples/CMakeLists.txt
+## XXX: In some versions of CMake clip wasn't being built before llama.
+## This is an hack for now, but it should be fixed in the future.
+	cp -rfv llama.cpp/examples/llava/clip.h llama.cpp/examples/grpc-server/clip.h
+	cp -rfv llama.cpp/examples/llava/clip.cpp llama.cpp/examples/grpc-server/clip.cpp

 rebuild:
 	cp -rfv $(abspath ./)/CMakeLists.txt llama.cpp/examples/grpc-server/
 	cp -rfv $(abspath ./)/grpc-server.cpp llama.cpp/examples/grpc-server/
+	cp -rfv $(abspath ./)/json.hpp llama.cpp/examples/grpc-server/
 	rm -rf grpc-server
 	$(MAKE) grpc-server

--- a/backend/cpp/llama/grpc-server.cpp
+++ b/backend/cpp/llama/grpc-server.cpp
--- a/backend/cpp/llama/json.hpp
+++ b/backend/cpp/llama/json.hpp
--- a/custom-ca-certs/.keep
+++ b/custom-ca-certs/.keep
--- a/entrypoint.sh
+++ b/entrypoint.sh
@@ -5,7 +5,7 @@ cd /build

 if [ "$REBUILD" != "false" ]; then
 	rm -rf ./local-ai
-	ESPEAK_DATA=/build/lib/Linux-$(uname -m)/piper_phonemize/lib/espeak-ng-data make build -j${BUILD_PARALLELISM:-1}
+	make build -j${BUILD_PARALLELISM:-1}
 else
 	echo "@@@@@"
 	echo "Skipping rebuild"
--- a/examples/chainlit/Dockerfile
+++ b/examples/chainlit/Dockerfile
@@ -0,0 +1,16 @@
+# Use an official Python runtime as a parent image
+FROM harbor.home.sfxworks.net/docker/library/python:3.9-slim
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Copy the current directory contents into the container at /app
+COPY requirements.txt /app
+
+# Install any needed packages specified in requirements.txt
+RUN pip install -r requirements.txt
+
+COPY . /app
+
+# Run app.py when the container launches
+CMD ["chainlit", "run", "-h", "--host", "0.0.0.0", "main.py" ]
--- a/examples/chainlit/README.md
+++ b/examples/chainlit/README.md
@@ -0,0 +1,25 @@
+# LocalAI Demonstration with Embeddings and Chainlit
+
+This demonstration shows you how to use embeddings with existing data in `LocalAI`, and how to integrate it with Chainlit for an interactive querying experience. We are using the `llama_index` library to facilitate the embedding and querying processes, and `chainlit` to provide an interactive interface. The `Weaviate` client is used as the embedding source.
+
+## Prerequisites
+
+Before proceeding, make sure you have the following installed:
+- Weaviate client
+- LocalAI and its dependencies
+- Chainlit and its dependencies
+
+## Getting Started
+
+1. Clone this repository:
+2. Navigate to the project directory:
+3. Run the example: `chainlit run main.py`
+
+# Highlight on `llama_index` and `chainlit`
+
+`llama_index` is the key library that facilitates the process of embedding and querying data in LocalAI. It provides a seamless interface to integrate various components, such as `WeaviateVectorStore`, `LocalAI`, `ServiceContext`, and more, for a smooth querying experience.
+
+`chainlit` is used to provide an interactive interface for users to query the data and see the results in real-time. It integrates with llama_index to handle the querying process and display the results to the user.
+
+In this example, `llama_index` is used to set up the `VectorStoreIndex` and `QueryEngine`, and `chainlit` is used to handle the user interactions with `LocalAI` and display the results.
+
--- a/examples/chainlit/config.yaml
+++ b/examples/chainlit/config.yaml
@@ -0,0 +1,16 @@
+localAI:
+  temperature: 0
+  modelName: gpt-3.5-turbo
+  apiBase: http://local-ai.default
+  apiKey: stub
+  streaming: True
+weviate:
+  url: http://weviate.local
+  index: AIChroma
+query:
+  mode: hybrid
+  topK: 1
+  alpha: 0.0
+  chunkSize: 1024
+embedding:
+  model: BAAI/bge-small-en-v1.5
--- a/examples/chainlit/main.py
+++ b/examples/chainlit/main.py
@@ -0,0 +1,82 @@
+import os
+
+import weaviate
+from llama_index.storage.storage_context import StorageContext
+from llama_index.vector_stores import WeaviateVectorStore
+
+from llama_index.query_engine.retriever_query_engine import RetrieverQueryEngine
+from llama_index.callbacks.base import CallbackManager
+from llama_index import (
+    LLMPredictor,
+    ServiceContext,
+    StorageContext,
+    VectorStoreIndex,
+)
+import chainlit as cl
+
+from llama_index.llms import LocalAI
+from llama_index.embeddings import HuggingFaceEmbedding
+import yaml
+
+# Load the configuration file
+with open("config.yaml", "r") as ymlfile:
+    cfg = yaml.safe_load(ymlfile)
+
+# Get the values from the configuration file or set the default values
+temperature = cfg['localAI'].get('temperature', 0)
+model_name = cfg['localAI'].get('modelName', "gpt-3.5-turbo")
+api_base = cfg['localAI'].get('apiBase', "http://local-ai.default")
+api_key = cfg['localAI'].get('apiKey', "stub")
+streaming = cfg['localAI'].get('streaming', True)
+weaviate_url = cfg['weviate'].get('url', "http://weviate.default")
+index_name = cfg['weviate'].get('index', "AIChroma")
+query_mode = cfg['query'].get('mode', "hybrid")
+topK = cfg['query'].get('topK', 1)
+alpha = cfg['query'].get('alpha', 0.0)
+embed_model_name = cfg['embedding'].get('model', "BAAI/bge-small-en-v1.5")
+chunk_size = cfg['query'].get('chunkSize', 1024)
+
+
+embed_model = HuggingFaceEmbedding(model_name=embed_model_name)
+
+
+llm = LocalAI(temperature=temperature, model_name=model_name, api_base=api_base, api_key=api_key, streaming=streaming)
+llm.globally_use_chat_completions = True;
+client = weaviate.Client(weaviate_url)
+vector_store = WeaviateVectorStore(weaviate_client=client, index_name=index_name)
+storage_context = StorageContext.from_defaults(vector_store=vector_store)
+
+@cl.on_chat_start
+async def factory():
+
+    llm_predictor = LLMPredictor(
+        llm=llm
+    )
+    
+    service_context = ServiceContext.from_defaults(embed_model=embed_model, callback_manager=CallbackManager([cl.LlamaIndexCallbackHandler()]), llm_predictor=llm_predictor, chunk_size=chunk_size)
+
+    index = VectorStoreIndex.from_vector_store(
+        vector_store,
+        storage_context=storage_context,
+        service_context=service_context
+    )
+
+    query_engine = index.as_query_engine(vector_store_query_mode=query_mode, similarity_top_k=topK, alpha=alpha, streaming=True)
+
+    cl.user_session.set("query_engine", query_engine)
+
+
+@cl.on_message
+async def main(message: cl.Message):
+    query_engine = cl.user_session.get("query_engine")
+    response = await cl.make_async(query_engine.query)(message.content)
+
+    response_message = cl.Message(content="")
+
+    for token in response.response_gen:
+        await response_message.stream_token(token=token)
+
+    if response.response_txt:
+        response_message.content = response.response_txt
+
+    await response_message.send()
--- a/examples/chainlit/requirements.txt
+++ b/examples/chainlit/requirements.txt
@@ -0,0 +1,7 @@
+llama_hub==0.0.41
+llama_index==0.8.55
+Requests==2.31.0
+weaviate_client==3.25.1
+transformers
+torch
+chainlit
--- a/examples/configurations/llava/README.md
+++ b/examples/configurations/llava/README.md
@@ -0,0 +1,18 @@
+![llava](https://github.com/mudler/LocalAI/assets/2420543/cb0a0897-3b58-4350-af66-e6f4387b58d3)
+
+## Setup
+
+```
+mkdir models
+wget https://huggingface.co/mys/ggml_bakllava-1/resolve/main/ggml-model-q4_k.gguf -O models/ggml-model-q4_k.gguf
+wget https://huggingface.co/mys/ggml_bakllava-1/resolve/main/mmproj-model-f16.gguf -O models/mmproj-model-f16.gguf
+docker run -p 8080:8080 -v $PWD/models:/models -ti --rm quay.io/go-skynet/local-ai:master --models-path /models --threads 4
+```
+
+## Try it out
+
+```
+curl http://localhost:8080/v1/chat/completions -H "Content-Type: application/json" -d '{
+     "model": "llava",
+     "messages": [{"role": "user", "content": [{"type":"text", "text": "What is in the image?"}, {"type": "image_url", "image_url": {"url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg" }}], "temperature": 0.9}]}'
+```
--- a/examples/configurations/llava/chat-simple.tmpl
+++ b/examples/configurations/llava/chat-simple.tmpl
@@ -0,0 +1,3 @@
+A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.
+{{.Input}}
+ASSISTANT:
--- a/examples/configurations/llava/llava.yaml
+++ b/examples/configurations/llava/llava.yaml
@@ -0,0 +1,20 @@
+
+context_size: 4096
+f16: true
+threads: 11
+gpu_layers: 90
+name: llava
+mmap: true
+backend: llama-cpp
+roles:
+  user: "USER:"
+  assistant: "ASSISTANT:"
+  system: "SYSTEM:"
+parameters:
+  model: ggml-model-q4_k.gguf
+  temperature: 0.2
+  top_k: 40
+  top_p: 0.95
+template:
+  chat: chat-simple
+mmproj: mmproj-model-f16.gguf
--- a/examples/llamaindex/README.md
+++ b/examples/llamaindex/README.md
@@ -0,0 +1,30 @@
+# LocalAI Demonstration with Embeddings
+
+This demonstration shows you how to use embeddings with existing data in LocalAI. We are using the `llama_index` library to facilitate the embedding and querying processes. The `Weaviate` client is used as the embedding source.
+
+## Prerequisites
+
+Before proceeding, make sure you have the following installed:
+- Weaviate client
+- LocalAI and its dependencies
+- llama_index and its dependencies 
+
+## Getting Started
+
+1. Clone this repository:
+
+2. Navigate to the project directory:
+
+3. Run the example:
+
+`python main.py`
+
+```
+Downloading (…)lve/main/config.json: 100%|███████████████████████████| 684/684 [00:00<00:00, 6.01MB/s]
+Downloading model.safetensors: 100%|███████████████████████████████| 133M/133M [00:03<00:00, 39.5MB/s]
+Downloading (…)okenizer_config.json: 100%|███████████████████████████| 366/366 [00:00<00:00, 2.79MB/s]
+Downloading (…)solve/main/vocab.txt: 100%|█████████████████████████| 232k/232k [00:00<00:00, 6.00MB/s]
+Downloading (…)/main/tokenizer.json: 100%|█████████████████████████| 711k/711k [00:00<00:00, 18.8MB/s]
+Downloading (…)cial_tokens_map.json: 100%|███████████████████████████| 125/125 [00:00<00:00, 1.18MB/s]
+LocalAI is a community-driven project that aims to make AI accessible to everyone. It was created by Ettore Di Giacinto and is focused on providing various AI-related features such as text generation with GPTs, text to audio, audio to text, image generation, and more. The project is constantly growing and evolving, with a roadmap for future improvements. Anyone is welcome to contribute, provide feedback, and submit pull requests to help make LocalAI better.
+```
--- a/examples/llamaindex/main.py
+++ b/examples/llamaindex/main.py
@@ -0,0 +1,38 @@
+import os
+
+import weaviate
+
+from llama_index import ServiceContext, VectorStoreIndex, StorageContext
+from llama_index.llms import LocalAI
+from llama_index.vector_stores import WeaviateVectorStore
+from llama_index.storage.storage_context import StorageContext
+
+# Weaviate client setup
+client = weaviate.Client("http://weviate.default")
+
+# Weaviate vector store setup
+vector_store = WeaviateVectorStore(weaviate_client=client, index_name="AIChroma")
+
+# Storage context setup
+storage_context = StorageContext.from_defaults(vector_store=vector_store)
+
+# LocalAI setup
+llm = LocalAI(temperature=0, model_name="gpt-3.5-turbo", api_base="http://local-ai.default", api_key="stub")
+llm.globally_use_chat_completions = True;
+
+# Service context setup
+service_context = ServiceContext.from_defaults(llm=llm, embed_model="local")
+
+# Load index from stored vectors
+index = VectorStoreIndex.from_vector_store(
+    vector_store,
+    storage_context=storage_context,
+    service_context=service_context
+)
+
+# Query engine setup
+query_engine = index.as_query_engine(similarity_top_k=1, vector_store_query_mode="hybrid")
+
+# Query example
+response = query_engine.query("What is LocalAI?")
+print(response)
--- a/extra/grpc/README.md
+++ b/extra/grpc/README.md
@@ -0,0 +1,38 @@
+# Common commands about conda environment
+
+## Create a new empty conda environment
+
+```
+conda create --name <env-name> python=<your version> -y
+
+conda create --name autogptq python=3.11 -y
+```
+
+## To activate the environment
+
+As of conda 4.4
+```
+conda activate autogptq
+```
+
+The conda version older than 4.4
+
+```
+source activate autogptq
+```
+
+## Install the packages to your environment
+
+Sometimes you need to install the packages from the conda-forge channel
+
+By using `conda`
+```
+conda install <your-package-name>
+
+conda install -c conda-forge <your package-name>
+```
+
+Or by using `pip`
+```
+pip install <your-package-name>
+```
--- a/extra/grpc/autogptq/Makefile
+++ b/extra/grpc/autogptq/Makefile
@@ -0,0 +1,5 @@
+.PONY: autogptq
+autogptq:
+	@echo "Creating virtual environment..."
+	@conda env create --name autogptq --file autogptq.yml
+	@echo "Virtual environment created."
--- a/extra/grpc/autogptq/README.md
+++ b/extra/grpc/autogptq/README.md
@@ -0,0 +1,5 @@
+# Creating a separate environment for the autogptq project
+
+```
+make autogptq
+```
--- a/extra/grpc/autogptq/autogptq.py
+++ b/extra/grpc/autogptq/autogptq.py
@@ -1,15 +1,15 @@
 #!/usr/bin/env python3
-import grpc
 from concurrent import futures
-import time
-import backend_pb2
-import backend_pb2_grpc
 import argparse
 import signal
 import sys
 import os
-from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
-from pathlib import Path
+import time
+
+import grpc
+import backend_pb2
+import backend_pb2_grpc
+from auto_gptq import AutoGPTQForCausalLM
 from transformers import AutoTokenizer
 from transformers import TextGenerationPipeline

--- a/extra/grpc/autogptq/autogptq.yml
+++ b/extra/grpc/autogptq/autogptq.yml
@@ -0,0 +1,86 @@
+name: autogptq
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - bzip2=1.0.8=h7b6447c_0
+  - ca-certificates=2023.08.22=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_0
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - libuuid=1.41.5=h5eee18b_0
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.11=h7f8727e_2
+  - pip=23.2.1=py311h06a4308_0
+  - python=3.11.5=h955ad1f_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=68.0.0=py311h06a4308_0
+  - sqlite=3.41.2=h5eee18b_0
+  - tk=8.6.12=h1ccaba5_0
+  - wheel=0.41.2=py311h06a4308_0
+  - xz=5.4.2=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+      - accelerate==0.23.0
+      - aiohttp==3.8.5
+      - aiosignal==1.3.1
+      - async-timeout==4.0.3
+      - attrs==23.1.0
+      - auto-gptq==0.4.2
+      - certifi==2023.7.22
+      - charset-normalizer==3.3.0
+      - datasets==2.14.5
+      - dill==0.3.7
+      - filelock==3.12.4
+      - frozenlist==1.4.0
+      - fsspec==2023.6.0
+      - grpcio==1.59.0
+      - huggingface-hub==0.16.4
+      - idna==3.4
+      - jinja2==3.1.2
+      - markupsafe==2.1.3
+      - mpmath==1.3.0
+      - multidict==6.0.4
+      - multiprocess==0.70.15
+      - networkx==3.1
+      - numpy==1.26.0
+      - nvidia-cublas-cu12==12.1.3.1
+      - nvidia-cuda-cupti-cu12==12.1.105
+      - nvidia-cuda-nvrtc-cu12==12.1.105
+      - nvidia-cuda-runtime-cu12==12.1.105
+      - nvidia-cudnn-cu12==8.9.2.26
+      - nvidia-cufft-cu12==11.0.2.54
+      - nvidia-curand-cu12==10.3.2.106
+      - nvidia-cusolver-cu12==11.4.5.107
+      - nvidia-cusparse-cu12==12.1.0.106
+      - nvidia-nccl-cu12==2.18.1
+      - nvidia-nvjitlink-cu12==12.2.140
+      - nvidia-nvtx-cu12==12.1.105
+      - packaging==23.2
+      - pandas==2.1.1
+      - peft==0.5.0
+      - protobuf==4.24.4
+      - psutil==5.9.5
+      - pyarrow==13.0.0
+      - python-dateutil==2.8.2
+      - pytz==2023.3.post1
+      - pyyaml==6.0.1
+      - regex==2023.10.3
+      - requests==2.31.0
+      - rouge==1.0.1
+      - safetensors==0.3.3
+      - six==1.16.0
+      - sympy==1.12
+      - tokenizers==0.14.0
+      - torch==2.1.0
+      - tqdm==4.66.1
+      - transformers==4.34.0
+      - triton==2.1.0
+      - typing-extensions==4.8.0
+      - tzdata==2023.3
+      - urllib3==2.0.6
+      - xxhash==3.4.1
+      - yarl==1.9.2
--- a/extra/grpc/autogptq/backend_pb2.py
+++ b/extra/grpc/autogptq/backend_pb2.py
--- a/extra/grpc/autogptq/run.sh
+++ b/extra/grpc/autogptq/run.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+
+##
+## A bash script wrapper that runs the autogptq server with conda
+
+export PATH=$PATH:/opt/conda/bin
+
+# Activate conda environment
+source activate autogptq
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python $DIR/autogptq.py $@
--- a/extra/grpc/bark/Makefile
+++ b/extra/grpc/bark/Makefile
@@ -0,0 +1,5 @@
+.PONY: ttsbark
+ttsbark:
+	@echo "Creating virtual environment..."
+	@conda env create --name ttsbark --file ttsbark.yml
+	@echo "Virtual environment created."
--- a/extra/grpc/bark/README.md
+++ b/extra/grpc/bark/README.md
@@ -0,0 +1,16 @@
+# Creating a separate environment for ttsbark project
+
+```
+make ttsbark
+```
+
+# Testing the gRPC server
+
+```
+<The path of your python interpreter> -m unittest test_ttsbark.py
+```
+
+For example
+```
+/opt/conda/envs/bark/bin/python -m unittest extra/grpc/bark/test_ttsbark.py
+``````
--- a/extra/grpc/bark/backend_pb2.py
+++ b/extra/grpc/bark/backend_pb2.py
--- a/extra/grpc/bark/run.sh
+++ b/extra/grpc/bark/run.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+
+##
+## A bash script wrapper that runs the ttsbark server with conda
+
+export PATH=$PATH:/opt/conda/bin
+
+# Activate conda environment
+source activate ttsbark
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python $DIR/ttsbark.py $@
--- a/extra/grpc/bark/test_ttsbark.py
+++ b/extra/grpc/bark/test_ttsbark.py
@@ -0,0 +1,32 @@
+import unittest
+import subprocess
+import time
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc
+
+class TestBackendServicer(unittest.TestCase):
+    """
+    TestBackendServicer is the class that tests the gRPC service
+    """
+    def setUp(self):
+        self.service = subprocess.Popen(["python3", "ttsbark.py", "--addr", "localhost:50051"])
+
+    def tearDown(self) -> None:
+        self.service.terminate()
+        self.service.wait()
+
+    def test_server_startup(self):
+        time.sleep(2)
+        try:
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                response = stub.Health(backend_pb2.HealthMessage())
+                self.assertEqual(response.message, b'OK')
+        except Exception as err:
+            print(err)
+            self.fail("Server failed to start")
+        finally:
+            self.tearDown()
--- a/extra/grpc/bark/ttsbark.py
+++ b/extra/grpc/bark/ttsbark.py
@@ -1,18 +1,23 @@
+"""
+This is the extra gRPC server of LocalAI
+"""
+
 #!/usr/bin/env python3
-import grpc
 from concurrent import futures
 import time
-import backend_pb2
-import backend_pb2_grpc
 import argparse
 import signal
 import sys
 import os
-from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
-from pathlib import Path
-from bark import SAMPLE_RATE, generate_audio, preload_models
 from scipy.io.wavfile import write as write_wav

+import backend_pb2
+import backend_pb2_grpc
+from bark import SAMPLE_RATE, generate_audio, preload_models
+
+import grpc
+
+
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24

 # If MAX_WORKERS are specified in the environment use it, otherwise default to 1
@@ -20,6 +25,9 @@ MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))

 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
+    """
+    BackendServicer is the class that implements the gRPC service
+    """
    def Health(self, request, context):
        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
    def LoadModel(self, request, context):
@@ -83,4 +91,4 @@ if __name__ == "__main__":
    )
    args = parser.parse_args()

-    serve(args.addr)
+    serve(args.addr)
--- a/extra/grpc/bark/ttsbark.yml
+++ b/extra/grpc/bark/ttsbark.yml
@@ -0,0 +1,96 @@
+name: bark
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - bzip2=1.0.8=h7b6447c_0
+  - ca-certificates=2023.08.22=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_0
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - libuuid=1.41.5=h5eee18b_0
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.11=h7f8727e_2
+  - pip=23.2.1=py311h06a4308_0
+  - python=3.11.5=h955ad1f_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=68.0.0=py311h06a4308_0
+  - sqlite=3.41.2=h5eee18b_0
+  - tk=8.6.12=h1ccaba5_0
+  - wheel=0.41.2=py311h06a4308_0
+  - xz=5.4.2=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+      - accelerate==0.23.0
+      - aiohttp==3.8.5
+      - aiosignal==1.3.1
+      - async-timeout==4.0.3
+      - attrs==23.1.0
+      - bark==0.1.5
+      - boto3==1.28.61
+      - botocore==1.31.61
+      - certifi==2023.7.22
+      - charset-normalizer==3.3.0
+      - datasets==2.14.5
+      - dill==0.3.7
+      - einops==0.7.0
+      - encodec==0.1.1
+      - filelock==3.12.4
+      - frozenlist==1.4.0
+      - fsspec==2023.6.0
+      - funcy==2.0
+      - grpcio==1.59.0
+      - huggingface-hub==0.16.4
+      - idna==3.4
+      - jinja2==3.1.2
+      - jmespath==1.0.1
+      - markupsafe==2.1.3
+      - mpmath==1.3.0
+      - multidict==6.0.4
+      - multiprocess==0.70.15
+      - networkx==3.1
+      - numpy==1.26.0
+      - nvidia-cublas-cu12==12.1.3.1
+      - nvidia-cuda-cupti-cu12==12.1.105
+      - nvidia-cuda-nvrtc-cu12==12.1.105
+      - nvidia-cuda-runtime-cu12==12.1.105
+      - nvidia-cudnn-cu12==8.9.2.26
+      - nvidia-cufft-cu12==11.0.2.54
+      - nvidia-curand-cu12==10.3.2.106
+      - nvidia-cusolver-cu12==11.4.5.107
+      - nvidia-cusparse-cu12==12.1.0.106
+      - nvidia-nccl-cu12==2.18.1
+      - nvidia-nvjitlink-cu12==12.2.140
+      - nvidia-nvtx-cu12==12.1.105
+      - packaging==23.2
+      - pandas==2.1.1
+      - peft==0.5.0
+      - protobuf==4.24.4
+      - psutil==5.9.5
+      - pyarrow==13.0.0
+      - python-dateutil==2.8.2
+      - pytz==2023.3.post1
+      - pyyaml==6.0.1
+      - regex==2023.10.3
+      - requests==2.31.0
+      - rouge==1.0.1
+      - s3transfer==0.7.0
+      - safetensors==0.3.3
+      - scipy==1.11.3
+      - six==1.16.0
+      - sympy==1.12
+      - tokenizers==0.14.0
+      - torch==2.1.0
+      - torchaudio==2.1.0
+      - tqdm==4.66.1
+      - transformers==4.34.0
+      - triton==2.1.0
+      - typing-extensions==4.8.0
+      - tzdata==2023.3
+      - urllib3==1.26.17
+      - xxhash==3.4.1
+      - yarl==1.9.2
+prefix: /opt/conda/envs/bark
--- a/extra/grpc/diffusers/Makefile
+++ b/extra/grpc/diffusers/Makefile
@@ -0,0 +1,11 @@
+.PONY: diffusers
+diffusers:
+	@echo "Creating virtual environment..."
+	@conda env create --name diffusers --file diffusers.yml
+	@echo "Virtual environment created."
+
+.PONY: run
+run:
+	@echo "Running diffusers..."
+	bash run.sh
+	@echo "Diffusers run."
--- a/extra/grpc/diffusers/README.md
+++ b/extra/grpc/diffusers/README.md
@@ -0,0 +1,5 @@
+# Creating a separate environment for the diffusers project
+
+```
+make diffusers
+```
--- a/extra/grpc/diffusers/backend_diffusers.py
+++ b/extra/grpc/diffusers/backend_diffusers.py
@@ -1,27 +1,32 @@
 #!/usr/bin/env python3
-import grpc
 from concurrent import futures
-import time
-import backend_pb2
-import backend_pb2_grpc
+
 import argparse
+from collections import defaultdict
+from enum import Enum
 import signal
 import sys
+import time
 import os

-# import diffusers
-import torch
-from torch import autocast
-from diffusers import StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, EulerAncestralDiscreteScheduler
-from diffusers.pipelines.stable_diffusion import safety_checker
-from compel import Compel
 from PIL import Image
-from io import BytesIO
+import torch
+
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc
+
+from diffusers import StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, EulerAncestralDiscreteScheduler
 from diffusers import StableDiffusionImg2ImgPipeline
+from diffusers.pipelines.stable_diffusion import safety_checker
+
+from compel import Compel
+
 from transformers import CLIPTextModel
-from enum import Enum
-from collections import defaultdict
 from safetensors.torch import load_file
+
+
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 COMPEL=os.environ.get("COMPEL", "1") == "1"
 CLIPSKIP=os.environ.get("CLIPSKIP", "1") == "1"
--- a/extra/grpc/diffusers/backend_pb2.py
+++ b/extra/grpc/diffusers/backend_pb2.py
--- a/extra/grpc/diffusers/diffusers.yml
+++ b/extra/grpc/diffusers/diffusers.yml
@@ -0,0 +1,74 @@
+name: diffusers
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - bzip2=1.0.8=h7b6447c_0
+  - ca-certificates=2023.08.22=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_0
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - libuuid=1.41.5=h5eee18b_0
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.11=h7f8727e_2
+  - pip=23.2.1=py311h06a4308_0
+  - python=3.11.5=h955ad1f_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=68.0.0=py311h06a4308_0
+  - sqlite=3.41.2=h5eee18b_0
+  - tk=8.6.12=h1ccaba5_0
+  - tzdata=2023c=h04d1e81_0
+  - wheel=0.41.2=py311h06a4308_0
+  - xz=5.4.2=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+      - accelerate==0.23.0
+      - certifi==2023.7.22
+      - charset-normalizer==3.3.0
+      - compel==2.0.2
+      - diffusers==0.21.4
+      - filelock==3.12.4
+      - fsspec==2023.9.2
+      - grpcio==1.59.0
+      - huggingface-hub==0.17.3
+      - idna==3.4
+      - importlib-metadata==6.8.0
+      - jinja2==3.1.2
+      - markupsafe==2.1.3
+      - mpmath==1.3.0
+      - networkx==3.1
+      - numpy==1.26.0
+      - nvidia-cublas-cu12==12.1.3.1
+      - nvidia-cuda-cupti-cu12==12.1.105
+      - nvidia-cuda-nvrtc-cu12==12.1.105
+      - nvidia-cuda-runtime-cu12==12.1.105
+      - nvidia-cudnn-cu12==8.9.2.26
+      - nvidia-cufft-cu12==11.0.2.54
+      - nvidia-curand-cu12==10.3.2.106
+      - nvidia-cusolver-cu12==11.4.5.107
+      - nvidia-cusparse-cu12==12.1.0.106
+      - nvidia-nccl-cu12==2.18.1
+      - nvidia-nvjitlink-cu12==12.2.140
+      - nvidia-nvtx-cu12==12.1.105
+      - packaging==23.2
+      - pillow==10.0.1
+      - protobuf==4.24.4
+      - psutil==5.9.5
+      - pyparsing==3.1.1
+      - pyyaml==6.0.1
+      - regex==2023.10.3
+      - requests==2.31.0
+      - safetensors==0.4.0
+      - sympy==1.12
+      - tokenizers==0.14.1
+      - torch==2.1.0
+      - tqdm==4.66.1
+      - transformers==4.34.0
+      - triton==2.1.0
+      - typing-extensions==4.8.0
+      - urllib3==2.0.6
+      - zipp==3.17.0
+prefix: /opt/conda/envs/diffusers
--- a/extra/grpc/diffusers/run.sh
+++ b/extra/grpc/diffusers/run.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+
+##
+## A bash script wrapper that runs the diffusers server with conda
+
+export PATH=$PATH:/opt/conda/bin
+
+# Activate conda environment
+source activate diffusers
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python $DIR/backend_diffusers.py $@
--- a/extra/grpc/exllama/Makefile
+++ b/extra/grpc/exllama/Makefile
@@ -0,0 +1,11 @@
+.PONY: exllama
+exllama:
+	@echo "Creating virtual environment..."
+	@conda env create --name exllama --file exllama.yml
+	@echo "Virtual environment created."
+
+.PONY: run
+run:
+	@echo "Running exllama..."
+	bash run.sh
+	@echo "exllama run."
--- a/extra/grpc/exllama/README.md
+++ b/extra/grpc/exllama/README.md
@@ -0,0 +1,5 @@
+# Creating a separate environment for the exllama project
+
+```
+make exllama
+```
--- a/extra/grpc/exllama/backend_pb2.py
+++ b/extra/grpc/exllama/backend_pb2.py
--- a/extra/grpc/exllama/exllama.yml
+++ b/extra/grpc/exllama/exllama.yml
@@ -0,0 +1,55 @@
+name: exllama
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - bzip2=1.0.8=h7b6447c_0
+  - ca-certificates=2023.08.22=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_0
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - libuuid=1.41.5=h5eee18b_0
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.11=h7f8727e_2
+  - pip=23.2.1=py311h06a4308_0
+  - python=3.11.5=h955ad1f_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=68.0.0=py311h06a4308_0
+  - sqlite=3.41.2=h5eee18b_0
+  - tk=8.6.12=h1ccaba5_0
+  - tzdata=2023c=h04d1e81_0
+  - wheel=0.41.2=py311h06a4308_0
+  - xz=5.4.2=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+      - filelock==3.12.4
+      - fsspec==2023.9.2
+      - grpcio==1.59.0
+      - jinja2==3.1.2
+      - markupsafe==2.1.3
+      - mpmath==1.3.0
+      - networkx==3.1
+      - ninja==1.11.1
+      - nvidia-cublas-cu12==12.1.3.1
+      - nvidia-cuda-cupti-cu12==12.1.105
+      - nvidia-cuda-nvrtc-cu12==12.1.105
+      - nvidia-cuda-runtime-cu12==12.1.105
+      - nvidia-cudnn-cu12==8.9.2.26
+      - nvidia-cufft-cu12==11.0.2.54
+      - nvidia-curand-cu12==10.3.2.106
+      - nvidia-cusolver-cu12==11.4.5.107
+      - nvidia-cusparse-cu12==12.1.0.106
+      - nvidia-nccl-cu12==2.18.1
+      - nvidia-nvjitlink-cu12==12.2.140
+      - nvidia-nvtx-cu12==12.1.105
+      - protobuf==4.24.4
+      - safetensors==0.3.2
+      - sentencepiece==0.1.99
+      - sympy==1.12
+      - torch==2.1.0
+      - triton==2.1.0
+      - typing-extensions==4.8.0
+prefix: /opt/conda/envs/exllama
--- a/extra/grpc/exllama/run.sh
+++ b/extra/grpc/exllama/run.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+
+##
+## A bash script wrapper that runs the exllama server with conda
+
+export PATH=$PATH:/opt/conda/bin
+
+# Activate conda environment
+source activate exllama
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python $DIR/exllama.py $@
--- a/extra/grpc/huggingface/Makefile
+++ b/extra/grpc/huggingface/Makefile
@@ -0,0 +1,18 @@
+.PONY: huggingface
+huggingface:
+	@echo "Creating virtual environment..."
+	@conda env create --name huggingface --file huggingface.yml
+	@echo "Virtual environment created."
+
+.PONY: run
+run:
+	@echo "Running huggingface..."
+	bash run.sh
+	@echo "huggingface run."
+
+# It is not working well by using command line. It only6 works with IDE like VSCode.
+.PONY: test
+test:
+	@echo "Testing huggingface..."
+	bash test.sh
+	@echo "huggingface tested."
--- a/extra/grpc/huggingface/README.md
+++ b/extra/grpc/huggingface/README.md
@@ -0,0 +1,5 @@
+# Creating a separate environment for the huggingface project
+
+```
+make huggingface
+```
--- a/extra/grpc/huggingface/backend_pb2.py
+++ b/extra/grpc/huggingface/backend_pb2.py
--- a/extra/grpc/huggingface/huggingface.py
+++ b/extra/grpc/huggingface/huggingface.py
@@ -1,13 +1,20 @@
+"""
+Extra gRPC server for HuggingFace SentenceTransformer models.
+"""
 #!/usr/bin/env python3
-import grpc
 from concurrent import futures
-import time
-import backend_pb2
-import backend_pb2_grpc
+
 import argparse
 import signal
 import sys
 import os
+
+import time
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc
+
 from sentence_transformers import SentenceTransformer

 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
@@ -17,18 +24,56 @@ MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))

 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
+    """
+    A gRPC servicer for the backend service.
+
+    This class implements the gRPC methods for the backend service, including Health, LoadModel, and Embedding.
+    """
    def Health(self, request, context):
+        """
+        A gRPC method that returns the health status of the backend service.
+
+        Args:
+            request: A HealthRequest object that contains the request parameters.
+            context: A grpc.ServicerContext object that provides information about the RPC.
+
+        Returns:
+            A Reply object that contains the health status of the backend service.
+        """
        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
+
    def LoadModel(self, request, context):
+        """
+        A gRPC method that loads a model into memory.
+
+        Args:
+            request: A LoadModelRequest object that contains the request parameters.
+            context: A grpc.ServicerContext object that provides information about the RPC.
+
+        Returns:
+            A Result object that contains the result of the LoadModel operation.
+        """
        model_name = request.Model
        try:
            self.model = SentenceTransformer(model_name)
        except Exception as err:
            return backend_pb2.Result(success=False, message=f"Unexpected {err=}, {type(err)=}")
+
        # Implement your logic here for the LoadModel service
        # Replace this with your desired response
        return backend_pb2.Result(message="Model loaded successfully", success=True)
+
    def Embedding(self, request, context):
+        """
+        A gRPC method that calculates embeddings for a given sentence.
+
+        Args:
+            request: An EmbeddingRequest object that contains the request parameters.
+            context: A grpc.ServicerContext object that provides information about the RPC.
+
+        Returns:
+            An EmbeddingResult object that contains the calculated embeddings.
+        """
        # Implement your logic here for the Embedding service
        # Replace this with your desired response
        print("Calculated embeddings for: " + request.Embeddings, file=sys.stderr)
@@ -66,4 +111,4 @@ if __name__ == "__main__":
    )
    args = parser.parse_args()

-    serve(args.addr)
+    serve(args.addr)
--- a/extra/grpc/huggingface/huggingface.yml
+++ b/extra/grpc/huggingface/huggingface.yml
@@ -0,0 +1,77 @@
+name: huggingface
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - bzip2=1.0.8=h7b6447c_0
+  - ca-certificates=2023.08.22=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_0
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - libuuid=1.41.5=h5eee18b_0
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.11=h7f8727e_2
+  - pip=23.2.1=py311h06a4308_0
+  - python=3.11.5=h955ad1f_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=68.0.0=py311h06a4308_0
+  - sqlite=3.41.2=h5eee18b_0
+  - tk=8.6.12=h1ccaba5_0
+  - tzdata=2023c=h04d1e81_0
+  - wheel=0.41.2=py311h06a4308_0
+  - xz=5.4.2=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+      - certifi==2023.7.22
+      - charset-normalizer==3.3.0
+      - click==8.1.7
+      - filelock==3.12.4
+      - fsspec==2023.9.2
+      - grpcio==1.59.0
+      - huggingface-hub==0.17.3
+      - idna==3.4
+      - install==1.3.5
+      - jinja2==3.1.2
+      - joblib==1.3.2
+      - markupsafe==2.1.3
+      - mpmath==1.3.0
+      - networkx==3.1
+      - nltk==3.8.1
+      - numpy==1.26.0
+      - nvidia-cublas-cu12==12.1.3.1
+      - nvidia-cuda-cupti-cu12==12.1.105
+      - nvidia-cuda-nvrtc-cu12==12.1.105
+      - nvidia-cuda-runtime-cu12==12.1.105
+      - nvidia-cudnn-cu12==8.9.2.26
+      - nvidia-cufft-cu12==11.0.2.54
+      - nvidia-curand-cu12==10.3.2.106
+      - nvidia-cusolver-cu12==11.4.5.107
+      - nvidia-cusparse-cu12==12.1.0.106
+      - nvidia-nccl-cu12==2.18.1
+      - nvidia-nvjitlink-cu12==12.2.140
+      - nvidia-nvtx-cu12==12.1.105
+      - packaging==23.2
+      - pillow==10.0.1
+      - protobuf==4.24.4
+      - pyyaml==6.0.1
+      - regex==2023.10.3
+      - requests==2.31.0
+      - safetensors==0.4.0
+      - scikit-learn==1.3.1
+      - scipy==1.11.3
+      - sentence-transformers==2.2.2
+      - sentencepiece==0.1.99
+      - sympy==1.12
+      - threadpoolctl==3.2.0
+      - tokenizers==0.14.1
+      - torch==2.1.0
+      - torchvision==0.16.0
+      - tqdm==4.66.1
+      - transformers==4.34.0
+      - triton==2.1.0
+      - typing-extensions==4.8.0
+      - urllib3==2.0.6
+prefix: /opt/conda/envs/huggingface
--- a/extra/grpc/huggingface/run.sh
+++ b/extra/grpc/huggingface/run.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+
+##
+## A bash script wrapper that runs the huggingface server with conda
+
+export PATH=$PATH:/opt/conda/bin
+
+# Activate conda environment
+source activate huggingface
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python $DIR/huggingface.py $@
--- a/extra/grpc/huggingface/test.sh
+++ b/extra/grpc/huggingface/test.sh
@@ -0,0 +1,11 @@
+#!/bin/bash
+##
+## A bash script wrapper that runs the huggingface server with conda
+
+# Activate conda environment
+source activate huggingface
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python -m unittest $DIR/test_huggingface.py
--- a/extra/grpc/huggingface/test_huggingface.py
+++ b/extra/grpc/huggingface/test_huggingface.py
@@ -0,0 +1,81 @@
+"""
+A test script to test the gRPC service
+"""
+import unittest
+import subprocess
+import time
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc
+
+
+class TestBackendServicer(unittest.TestCase):
+    """
+    TestBackendServicer is the class that tests the gRPC service
+    """
+    def setUp(self):
+        """
+        This method sets up the gRPC service by starting the server
+        """
+        self.service = subprocess.Popen(["python3", "huggingface.py", "--addr", "localhost:50051"])
+
+    def tearDown(self) -> None:
+        """
+        This method tears down the gRPC service by terminating the server
+        """
+        self.service.terminate()
+        self.service.wait()
+
+    def test_server_startup(self):
+        """
+        This method tests if the server starts up successfully
+        """
+        time.sleep(2)
+        try:
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                response = stub.Health(backend_pb2.HealthMessage())
+                self.assertEqual(response.message, b'OK')
+        except Exception as err:
+            print(err)
+            self.fail("Server failed to start")
+        finally:
+            self.tearDown()
+
+    def test_load_model(self):
+        """
+        This method tests if the model is loaded successfully
+        """
+        try:
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                response = stub.LoadModel(backend_pb2.ModelOptions(Model="bert-base-nli-mean-tokens"))
+                self.assertTrue(response.success)
+                self.assertEqual(response.message, "Model loaded successfully")
+        except Exception as err:
+            print(err)
+            self.fail("LoadModel service failed")
+        finally:
+            self.tearDown()
+
+    def test_embedding(self):
+        """
+        This method tests if the embeddings are generated successfully
+        """
+        try:
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                response = stub.LoadModel(backend_pb2.ModelOptions(Model="bert-base-nli-mean-tokens"))
+                self.assertTrue(response.success)
+                embedding_request = backend_pb2.PredictOptions(Embeddings="This is a test sentence.")
+                embedding_response = stub.Embedding(embedding_request)
+                self.assertIsNotNone(embedding_response.embeddings)
+        except Exception as err:
+            print(err)
+            self.fail("Embedding service failed")
+        finally:
+            self.tearDown()
--- a/extra/grpc/vall-e-x/Makefile
+++ b/extra/grpc/vall-e-x/Makefile
@@ -0,0 +1,11 @@
+.PONY: ttsvalle
+ttsvalle:
+	@echo "Creating virtual environment..."
+	@conda env create --name ttsvalle --file ttsvalle.yml
+	@echo "Virtual environment created."
+
+.PONY: run
+run:
+	@echo "Running ttsvalle..."
+	bash run.sh
+	@echo "ttsvalle run."
--- a/extra/grpc/vall-e-x/README.md
+++ b/extra/grpc/vall-e-x/README.md
@@ -0,0 +1,5 @@
+# Creating a separate environment for the ttsvalle project
+
+```
+make ttsvalle
+```
--- a/extra/grpc/vall-e-x/backend_pb2.py
+++ b/extra/grpc/vall-e-x/backend_pb2.py
--- a/extra/grpc/vall-e-x/run.sh
+++ b/extra/grpc/vall-e-x/run.sh
@@ -0,0 +1,13 @@
+#!/bin/bash
+
+##
+## A bash script wrapper that runs the ttsvalle server with conda
+export PATH=$PATH:/opt/conda/bin
+
+# Activate conda environment
+source activate ttsvalle
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python $DIR/ttvalle.py $@
--- a/extra/grpc/vall-e-x/ttsvalle.py
+++ b/extra/grpc/vall-e-x/ttsvalle.py
@@ -1,14 +1,15 @@
 #!/usr/bin/env python3
-import grpc
+
 from concurrent import futures
-import time
-import backend_pb2
-import backend_pb2_grpc
 import argparse
 import signal
 import sys
 import os
-from pathlib import Path
+import time
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc

 from utils.generation import SAMPLE_RATE, generate_audio, preload_models
 from scipy.io.wavfile import write as write_wav
@@ -21,9 +22,34 @@ MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))

 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
+    """
+    gRPC servicer for backend services.
+    """
    def Health(self, request, context):
+        """
+        Health check service.
+
+        Args:
+            request: A backend_pb2.HealthRequest instance.
+            context: A grpc.ServicerContext instance.
+
+        Returns:
+            A backend_pb2.Reply instance with message "OK".
+        """
        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
+
    def LoadModel(self, request, context):
+        """
+        Load model service.
+
+        Args:
+            request: A backend_pb2.LoadModelRequest instance.
+            context: A grpc.ServicerContext instance.
+
+        Returns:
+            A backend_pb2.Result instance with message "Model loaded successfully" and success=True if successful.
+            A backend_pb2.Result instance with success=False and error message if unsuccessful.
+        """
        model_name = request.Model
        try:
            print("Preparing models, please wait", file=sys.stderr)
@@ -49,6 +75,17 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
        return backend_pb2.Result(message="Model loaded successfully", success=True)

    def TTS(self, request, context):
+        """
+        Text-to-speech service.
+
+        Args:
+            request: A backend_pb2.TTSRequest instance.
+            context: A grpc.ServicerContext instance.
+
+        Returns:
+            A backend_pb2.Result instance with success=True if successful.
+            A backend_pb2.Result instance with success=False and error message if unsuccessful.
+        """
        model = request.model
        print(request, file=sys.stderr)
        try:
@@ -97,4 +134,4 @@ if __name__ == "__main__":
    )
    args = parser.parse_args()

-    serve(args.addr)
+    serve(args.addr)
--- a/extra/grpc/vall-e-x/ttsvalle.yml
+++ b/extra/grpc/vall-e-x/ttsvalle.yml
@@ -0,0 +1,101 @@
+name: ttsvalle
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - bzip2=1.0.8=h7b6447c_0
+  - ca-certificates=2023.08.22=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_0
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - libuuid=1.41.5=h5eee18b_0
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.11=h7f8727e_2
+  - pip=23.2.1=py310h06a4308_0
+  - python=3.10.13=h955ad1f_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=68.0.0=py310h06a4308_0
+  - sqlite=3.41.2=h5eee18b_0
+  - tk=8.6.12=h1ccaba5_0
+  - tzdata=2023c=h04d1e81_0
+  - wheel=0.41.2=py310h06a4308_0
+  - xz=5.4.2=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+      - aiofiles==23.2.1
+      - altair==5.1.2
+      - annotated-types==0.6.0
+      - anyio==3.7.1
+      - click==8.1.7
+      - cn2an==0.5.22
+      - cython==3.0.3
+      - einops==0.7.0
+      - encodec==0.1.1
+      - eng-to-ipa==0.0.2
+      - fastapi==0.103.2
+      - ffmpeg-python==0.2.0
+      - ffmpy==0.3.1
+      - fsspec==2023.9.2
+      - future==0.18.3
+      - gradio==3.47.1
+      - gradio-client==0.6.0
+      - grpcio==1.59.0
+      - h11==0.14.0
+      - httpcore==0.18.0
+      - httpx==0.25.0
+      - huggingface-hub==0.17.3
+      - importlib-resources==6.1.0
+      - inflect==7.0.0
+      - jieba==0.42.1
+      - langid==1.1.6
+      - llvmlite==0.41.0
+      - more-itertools==10.1.0
+      - nltk==3.8.1
+      - numba==0.58.0
+      - numpy==1.25.2
+      - nvidia-cublas-cu12==12.1.3.1
+      - nvidia-cuda-cupti-cu12==12.1.105
+      - nvidia-cuda-nvrtc-cu12==12.1.105
+      - nvidia-cuda-runtime-cu12==12.1.105
+      - nvidia-cudnn-cu12==8.9.2.26
+      - nvidia-cufft-cu12==11.0.2.54
+      - nvidia-curand-cu12==10.3.2.106
+      - nvidia-cusolver-cu12==11.4.5.107
+      - nvidia-cusparse-cu12==12.1.0.106
+      - nvidia-nccl-cu12==2.18.1
+      - nvidia-nvjitlink-cu12==12.2.140
+      - nvidia-nvtx-cu12==12.1.105
+      - openai-whisper==20230306
+      - orjson==3.9.7
+      - proces==0.1.7
+      - protobuf==4.24.4
+      - pydantic==2.4.2
+      - pydantic-core==2.10.1
+      - pydub==0.25.1
+      - pyopenjtalk-prebuilt==0.3.0
+      - pypinyin==0.49.0
+      - python-multipart==0.0.6
+      - regex==2023.10.3
+      - safetensors==0.4.0
+      - semantic-version==2.10.0
+      - soundfile==0.12.1
+      - starlette==0.27.0
+      - sudachidict-core==20230927
+      - sudachipy==0.6.7
+      - tokenizers==0.14.1
+      - toolz==0.12.0
+      - torch==2.1.0
+      - torchaudio==2.1.0
+      - torchvision==0.16.0
+      - tqdm==4.66.1
+      - transformers==4.34.0
+      - triton==2.1.0
+      - unidecode==1.3.7
+      - uvicorn==0.23.2
+      - vocos==0.0.3
+      - websockets==11.0.3
+      - wget==3.2
+prefix: /opt/conda/envs/ttsvalle
--- a/extra/grpc/vllm/Makefile
+++ b/extra/grpc/vllm/Makefile
@@ -0,0 +1,11 @@
+.PONY: vllm
+vllm:
+	@echo "Creating virtual environment..."
+	@conda env create --name vllm --file vllm.yml
+	@echo "Virtual environment created."
+
+.PONY: run
+run:
+	@echo "Running vllm..."
+	bash run.sh
+	@echo "vllm run."
--- a/extra/grpc/vllm/README.md
+++ b/extra/grpc/vllm/README.md
@@ -0,0 +1,5 @@
+# Creating a separate environment for the vllm project
+
+```
+make vllm
+```
--- a/extra/grpc/vllm/backend_pb2.py
+++ b/extra/grpc/vllm/backend_pb2.py
--- a/extra/grpc/vllm/backend_vllm.py
+++ b/extra/grpc/vllm/backend_vllm.py
@@ -1,15 +1,15 @@
 #!/usr/bin/env python3
-import grpc
 from concurrent import futures
 import time
-import backend_pb2
-import backend_pb2_grpc
 import argparse
 import signal
 import sys
-import os, glob
+import os

-from pathlib import Path
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc
 from vllm import LLM, SamplingParams

 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
@@ -19,7 +19,20 @@ MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))

 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
+    """
+    A gRPC servicer that implements the Backend service defined in backend.proto.
+    """
    def generate(self,prompt, max_new_tokens):
+        """
+        Generates text based on the given prompt and maximum number of new tokens.
+
+        Args:
+            prompt (str): The prompt to generate text from.
+            max_new_tokens (int): The maximum number of new tokens to generate.
+
+        Returns:
+            str: The generated text.
+        """
        self.generator.end_beam_search()

        # Tokenizing the input
@@ -41,9 +54,31 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
            if token.item() == self.generator.tokenizer.eos_token_id:
                break
        return decoded_text
+
    def Health(self, request, context):
+        """
+        Returns a health check message.
+
+        Args:
+            request: The health check request.
+            context: The gRPC context.
+
+        Returns:
+            backend_pb2.Reply: The health check reply.
+        """
        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
+
    def LoadModel(self, request, context):
+        """
+        Loads a language model.
+
+        Args:
+            request: The load model request.
+            context: The gRPC context.
+
+        Returns:
+            backend_pb2.Result: The load model result.
+        """
        try:
            if request.Quantization != "":
                self.llm = LLM(model=request.Model, quantization=request.Quantization)
@@ -54,6 +89,16 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
        return backend_pb2.Result(message="Model loaded successfully", success=True)

    def Predict(self, request, context):
+        """
+        Generates text based on the given prompt and sampling parameters.
+
+        Args:
+            request: The predict request.
+            context: The gRPC context.
+
+        Returns:
+            backend_pb2.Result: The predict result.
+        """
        if request.TopP == 0:
            request.TopP = 0.9

@@ -68,6 +113,16 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
        return backend_pb2.Result(message=bytes(generated_text, encoding='utf-8'))

    def PredictStream(self, request, context):
+        """
+        Generates text based on the given prompt and sampling parameters, and streams the results.
+
+        Args:
+            request: The predict stream request.
+            context: The gRPC context.
+
+        Returns:
+            backend_pb2.Result: The predict stream result.
+        """
        # Implement PredictStream RPC
        #for reply in some_data_generator():
        #    yield reply
@@ -104,4 +159,4 @@ if __name__ == "__main__":
    )
    args = parser.parse_args()

-    serve(args.addr)
+    serve(args.addr)
--- a/extra/grpc/vllm/run.sh
+++ b/extra/grpc/vllm/run.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+
+##
+## A bash script wrapper that runs the diffusers server with conda
+
+export PATH=$PATH:/opt/conda/bin
+
+# Activate conda environment
+source activate vllm
+
+# get the directory where the bash script is located
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+python $DIR/backend_vllm.py $@
--- a/extra/grpc/vllm/test_backend_vllm.py
+++ b/extra/grpc/vllm/test_backend_vllm.py
@@ -0,0 +1,41 @@
+import unittest
+import subprocess
+import time
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc
+
+import unittest
+import subprocess
+import time
+import grpc
+import backend_pb2_grpc
+import backend_pb2
+
+class TestBackendServicer(unittest.TestCase):
+    """
+    TestBackendServicer is the class that tests the gRPC service.
+
+    This class contains methods to test the startup and shutdown of the gRPC service.
+    """
+    def setUp(self):
+        self.service = subprocess.Popen(["python", "backend_vllm.py", "--addr", "localhost:50051"])
+
+    def tearDown(self) -> None:
+        self.service.terminate()
+        self.service.wait()
+
+    def test_server_startup(self):
+        time.sleep(2)
+        try:
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                response = stub.Health(backend_pb2.HealthMessage())
+                self.assertEqual(response.message, b'OK')
+        except Exception as err:
+            print(err)
+            self.fail("Server failed to start")
+        finally:
+            self.tearDown()
--- a/extra/grpc/vllm/vllm.yml
+++ b/extra/grpc/vllm/vllm.yml
@@ -0,0 +1,99 @@
+name: vllm
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - bzip2=1.0.8=h7b6447c_0
+  - ca-certificates=2023.08.22=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_0
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - libuuid=1.41.5=h5eee18b_0
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.11=h7f8727e_2
+  - pip=23.2.1=py311h06a4308_0
+  - python=3.11.5=h955ad1f_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=68.0.0=py311h06a4308_0
+  - sqlite=3.41.2=h5eee18b_0
+  - tk=8.6.12=h1ccaba5_0
+  - wheel=0.41.2=py311h06a4308_0
+  - xz=5.4.2=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+      - aiosignal==1.3.1
+      - anyio==3.7.1
+      - attrs==23.1.0
+      - certifi==2023.7.22
+      - charset-normalizer==3.3.0
+      - click==8.1.7
+      - cmake==3.27.6
+      - fastapi==0.103.2
+      - filelock==3.12.4
+      - frozenlist==1.4.0
+      - fsspec==2023.9.2
+      - grpcio==1.59.0
+      - h11==0.14.0
+      - httptools==0.6.0
+      - huggingface-hub==0.17.3
+      - idna==3.4
+      - jinja2==3.1.2
+      - jsonschema==4.19.1
+      - jsonschema-specifications==2023.7.1
+      - lit==17.0.2
+      - markupsafe==2.1.3
+      - mpmath==1.3.0
+      - msgpack==1.0.7
+      - networkx==3.1
+      - ninja==1.11.1
+      - numpy==1.26.0
+      - nvidia-cublas-cu11==11.10.3.66
+      - nvidia-cuda-cupti-cu11==11.7.101
+      - nvidia-cuda-nvrtc-cu11==11.7.99
+      - nvidia-cuda-runtime-cu11==11.7.99
+      - nvidia-cudnn-cu11==8.5.0.96
+      - nvidia-cufft-cu11==10.9.0.58
+      - nvidia-curand-cu11==10.2.10.91
+      - nvidia-cusolver-cu11==11.4.0.1
+      - nvidia-cusparse-cu11==11.7.4.91
+      - nvidia-nccl-cu11==2.14.3
+      - nvidia-nvtx-cu11==11.7.91
+      - packaging==23.2
+      - pandas==2.1.1
+      - protobuf==4.24.4
+      - psutil==5.9.5
+      - pyarrow==13.0.0
+      - pydantic==1.10.13
+      - python-dateutil==2.8.2
+      - python-dotenv==1.0.0
+      - pytz==2023.3.post1
+      - pyyaml==6.0.1
+      - ray==2.7.0
+      - referencing==0.30.2
+      - regex==2023.10.3
+      - requests==2.31.0
+      - rpds-py==0.10.4
+      - safetensors==0.4.0
+      - sentencepiece==0.1.99
+      - six==1.16.0
+      - sniffio==1.3.0
+      - starlette==0.27.0
+      - sympy==1.12
+      - tokenizers==0.14.1
+      - torch==2.0.1
+      - tqdm==4.66.1
+      - transformers==4.34.0
+      - triton==2.0.0
+      - typing-extensions==4.8.0
+      - tzdata==2023.3
+      - urllib3==2.0.6
+      - uvicorn==0.23.2
+      - uvloop==0.17.0
+      - vllm==0.2.0
+      - watchfiles==0.20.0
+      - websockets==11.0.3
+      - xformers==0.0.22
+prefix: /opt/conda/envs/vllm
--- a/pkg/grammar/json_schema.go
+++ b/pkg/grammar/json_schema.go
@@ -10,6 +10,33 @@ import (
 	"strings"
 )

+const (
+	JSONBNF = `root   ::= object
+value  ::= object | array | string | number | ("true" | "false" | "null") ws
+
+object ::=
+  "{" ws (
+            string ":" ws value
+    ("," ws string ":" ws value)*
+  )? "}" ws
+
+array  ::=
+  "[" ws (
+            value
+    ("," ws value)*
+  )? "]" ws
+
+string ::=
+  "\"" (
+    [^"\\] |
+    "\\" (["\\/bfnrt] | "u" [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F]) # escapes
+  )* "\"" ws
+
+number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? ([eE] [-+]? [0-9]+)? ws
+
+ws ::= ([ \t\n] ws)?`
+)
+
 var (
 	SPACE_RULE = `" "?`

--- a/pkg/grpc/proto/backend.pb.go
+++ b/pkg/grpc/proto/backend.pb.go
@@ -1,7 +1,7 @@
 // Code generated by protoc-gen-go. DO NOT EDIT.
 // versions:
 // 	protoc-gen-go v1.26.0
-// 	protoc        v3.15.8
+// 	protoc        v4.23.4
 // source: pkg/grpc/proto/backend.proto

 package proto
@@ -156,6 +156,7 @@ type PredictOptions struct {
 	NegativePromptScale float32  `protobuf:"fixed32,39,opt,name=NegativePromptScale,proto3" json:"NegativePromptScale,omitempty"`
 	NegativePrompt      string   `protobuf:"bytes,40,opt,name=NegativePrompt,proto3" json:"NegativePrompt,omitempty"`
 	NDraft              int32    `protobuf:"varint,41,opt,name=NDraft,proto3" json:"NDraft,omitempty"`
+	Images              []string `protobuf:"bytes,42,rep,name=Images,proto3" json:"Images,omitempty"`
 }

 func (x *PredictOptions) Reset() {
@@ -470,6 +471,13 @@ func (x *PredictOptions) GetNDraft() int32 {
 	return 0
 }

+func (x *PredictOptions) GetImages() []string {
+	if x != nil {
+		return x.Images
+	}
+	return nil
+}
+
 // The response message containing the result
 type Reply struct {
 	state         protoimpl.MessageState
@@ -561,12 +569,20 @@ type ModelOptions struct {
 	// RWKV
 	Tokenizer string `protobuf:"bytes,34,opt,name=Tokenizer,proto3" json:"Tokenizer,omitempty"`
 	// LLM (llama.cpp)
-	LoraBase     string `protobuf:"bytes,35,opt,name=LoraBase,proto3" json:"LoraBase,omitempty"`
-	LoraAdapter  string `protobuf:"bytes,36,opt,name=LoraAdapter,proto3" json:"LoraAdapter,omitempty"`
-	NoMulMatQ    bool   `protobuf:"varint,37,opt,name=NoMulMatQ,proto3" json:"NoMulMatQ,omitempty"`
-	DraftModel   string `protobuf:"bytes,39,opt,name=DraftModel,proto3" json:"DraftModel,omitempty"`
-	AudioPath    string `protobuf:"bytes,38,opt,name=AudioPath,proto3" json:"AudioPath,omitempty"`
-	Quantization string `protobuf:"bytes,40,opt,name=Quantization,proto3" json:"Quantization,omitempty"`
+	LoraBase    string  `protobuf:"bytes,35,opt,name=LoraBase,proto3" json:"LoraBase,omitempty"`
+	LoraAdapter string  `protobuf:"bytes,36,opt,name=LoraAdapter,proto3" json:"LoraAdapter,omitempty"`
+	LoraScale   float32 `protobuf:"fixed32,42,opt,name=LoraScale,proto3" json:"LoraScale,omitempty"`
+	NoMulMatQ   bool    `protobuf:"varint,37,opt,name=NoMulMatQ,proto3" json:"NoMulMatQ,omitempty"`
+	DraftModel  string  `protobuf:"bytes,39,opt,name=DraftModel,proto3" json:"DraftModel,omitempty"`
+	AudioPath   string  `protobuf:"bytes,38,opt,name=AudioPath,proto3" json:"AudioPath,omitempty"`
+	// vllm
+	Quantization   string  `protobuf:"bytes,40,opt,name=Quantization,proto3" json:"Quantization,omitempty"`
+	MMProj         string  `protobuf:"bytes,41,opt,name=MMProj,proto3" json:"MMProj,omitempty"`
+	RopeScaling    string  `protobuf:"bytes,43,opt,name=RopeScaling,proto3" json:"RopeScaling,omitempty"`
+	YarnExtFactor  float32 `protobuf:"fixed32,44,opt,name=YarnExtFactor,proto3" json:"YarnExtFactor,omitempty"`
+	YarnAttnFactor float32 `protobuf:"fixed32,45,opt,name=YarnAttnFactor,proto3" json:"YarnAttnFactor,omitempty"`
+	YarnBetaFast   float32 `protobuf:"fixed32,46,opt,name=YarnBetaFast,proto3" json:"YarnBetaFast,omitempty"`
+	YarnBetaSlow   float32 `protobuf:"fixed32,47,opt,name=YarnBetaSlow,proto3" json:"YarnBetaSlow,omitempty"`
 }

 func (x *ModelOptions) Reset() {
@@ -853,6 +869,13 @@ func (x *ModelOptions) GetLoraAdapter() string {
 	return ""
 }

+func (x *ModelOptions) GetLoraScale() float32 {
+	if x != nil {
+		return x.LoraScale
+	}
+	return 0
+}
+
 func (x *ModelOptions) GetNoMulMatQ() bool {
 	if x != nil {
 		return x.NoMulMatQ
@@ -881,6 +904,48 @@ func (x *ModelOptions) GetQuantization() string {
 	return ""
 }

+func (x *ModelOptions) GetMMProj() string {
+	if x != nil {
+		return x.MMProj
+	}
+	return ""
+}
+
+func (x *ModelOptions) GetRopeScaling() string {
+	if x != nil {
+		return x.RopeScaling
+	}
+	return ""
+}
+
+func (x *ModelOptions) GetYarnExtFactor() float32 {
+	if x != nil {
+		return x.YarnExtFactor
+	}
+	return 0
+}
+
+func (x *ModelOptions) GetYarnAttnFactor() float32 {
+	if x != nil {
+		return x.YarnAttnFactor
+	}
+	return 0
+}
+
+func (x *ModelOptions) GetYarnBetaFast() float32 {
+	if x != nil {
+		return x.YarnBetaFast
+	}
+	return 0
+}
+
+func (x *ModelOptions) GetYarnBetaSlow() float32 {
+	if x != nil {
+		return x.YarnBetaSlow
+	}
+	return 0
+}
+
 type Result struct {
 	state         protoimpl.MessageState
 	sizeCache     protoimpl.SizeCache
@@ -1542,7 +1607,7 @@ var file_pkg_grpc_proto_backend_proto_rawDesc = []byte{
 	0x0a, 0x1c, 0x70, 0x6b, 0x67, 0x2f, 0x67, 0x72, 0x70, 0x63, 0x2f, 0x70, 0x72, 0x6f, 0x74, 0x6f,
 	0x2f, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x12, 0x07,
 	0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x22, 0x0f, 0x0a, 0x0d, 0x48, 0x65, 0x61, 0x6c, 0x74,
-	0x68, 0x4d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x22, 0xdc, 0x09, 0x0a, 0x0e, 0x50, 0x72, 0x65,
+	0x68, 0x4d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x22, 0xf4, 0x09, 0x0a, 0x0e, 0x50, 0x72, 0x65,
 	0x64, 0x69, 0x63, 0x74, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x12, 0x16, 0x0a, 0x06, 0x50,
 	0x72, 0x6f, 0x6d, 0x70, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x06, 0x50, 0x72, 0x6f,
 	0x6d, 0x70, 0x74, 0x12, 0x12, 0x0a, 0x04, 0x53, 0x65, 0x65, 0x64, 0x18, 0x02, 0x20, 0x01, 0x28,
@@ -1620,209 +1685,226 @@ var file_pkg_grpc_proto_backend_proto_rawDesc = []byte{
 	0x69, 0x76, 0x65, 0x50, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x18, 0x28, 0x20, 0x01, 0x28, 0x09, 0x52,
 	0x0e, 0x4e, 0x65, 0x67, 0x61, 0x74, 0x69, 0x76, 0x65, 0x50, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x12,
 	0x16, 0x0a, 0x06, 0x4e, 0x44, 0x72, 0x61, 0x66, 0x74, 0x18, 0x29, 0x20, 0x01, 0x28, 0x05, 0x52,
-	0x06, 0x4e, 0x44, 0x72, 0x61, 0x66, 0x74, 0x22, 0x21, 0x0a, 0x05, 0x52, 0x65, 0x70, 0x6c, 0x79,
-	0x12, 0x18, 0x0a, 0x07, 0x6d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x18, 0x01, 0x20, 0x01, 0x28,
-	0x0c, 0x52, 0x07, 0x6d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x22, 0xbe, 0x09, 0x0a, 0x0c, 0x4d,
-	0x6f, 0x64, 0x65, 0x6c, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x12, 0x14, 0x0a, 0x05, 0x4d,
-	0x6f, 0x64, 0x65, 0x6c, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x05, 0x4d, 0x6f, 0x64, 0x65,
-	0x6c, 0x12, 0x20, 0x0a, 0x0b, 0x43, 0x6f, 0x6e, 0x74, 0x65, 0x78, 0x74, 0x53, 0x69, 0x7a, 0x65,
-	0x18, 0x02, 0x20, 0x01, 0x28, 0x05, 0x52, 0x0b, 0x43, 0x6f, 0x6e, 0x74, 0x65, 0x78, 0x74, 0x53,
-	0x69, 0x7a, 0x65, 0x12, 0x12, 0x0a, 0x04, 0x53, 0x65, 0x65, 0x64, 0x18, 0x03, 0x20, 0x01, 0x28,
-	0x05, 0x52, 0x04, 0x53, 0x65, 0x65, 0x64, 0x12, 0x16, 0x0a, 0x06, 0x4e, 0x42, 0x61, 0x74, 0x63,
-	0x68, 0x18, 0x04, 0x20, 0x01, 0x28, 0x05, 0x52, 0x06, 0x4e, 0x42, 0x61, 0x74, 0x63, 0x68, 0x12,
-	0x1c, 0x0a, 0x09, 0x46, 0x31, 0x36, 0x4d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x18, 0x05, 0x20, 0x01,
-	0x28, 0x08, 0x52, 0x09, 0x46, 0x31, 0x36, 0x4d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x12, 0x14, 0x0a,
-	0x05, 0x4d, 0x4c, 0x6f, 0x63, 0x6b, 0x18, 0x06, 0x20, 0x01, 0x28, 0x08, 0x52, 0x05, 0x4d, 0x4c,
-	0x6f, 0x63, 0x6b, 0x12, 0x12, 0x0a, 0x04, 0x4d, 0x4d, 0x61, 0x70, 0x18, 0x07, 0x20, 0x01, 0x28,
-	0x08, 0x52, 0x04, 0x4d, 0x4d, 0x61, 0x70, 0x12, 0x1c, 0x0a, 0x09, 0x56, 0x6f, 0x63, 0x61, 0x62,
-	0x4f, 0x6e, 0x6c, 0x79, 0x18, 0x08, 0x20, 0x01, 0x28, 0x08, 0x52, 0x09, 0x56, 0x6f, 0x63, 0x61,
-	0x62, 0x4f, 0x6e, 0x6c, 0x79, 0x12, 0x18, 0x0a, 0x07, 0x4c, 0x6f, 0x77, 0x56, 0x52, 0x41, 0x4d,
-	0x18, 0x09, 0x20, 0x01, 0x28, 0x08, 0x52, 0x07, 0x4c, 0x6f, 0x77, 0x56, 0x52, 0x41, 0x4d, 0x12,
-	0x1e, 0x0a, 0x0a, 0x45, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x73, 0x18, 0x0a, 0x20,
-	0x01, 0x28, 0x08, 0x52, 0x0a, 0x45, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x73, 0x12,
-	0x12, 0x0a, 0x04, 0x4e, 0x55, 0x4d, 0x41, 0x18, 0x0b, 0x20, 0x01, 0x28, 0x08, 0x52, 0x04, 0x4e,
-	0x55, 0x4d, 0x41, 0x12, 0x1e, 0x0a, 0x0a, 0x4e, 0x47, 0x50, 0x55, 0x4c, 0x61, 0x79, 0x65, 0x72,
-	0x73, 0x18, 0x0c, 0x20, 0x01, 0x28, 0x05, 0x52, 0x0a, 0x4e, 0x47, 0x50, 0x55, 0x4c, 0x61, 0x79,
-	0x65, 0x72, 0x73, 0x12, 0x18, 0x0a, 0x07, 0x4d, 0x61, 0x69, 0x6e, 0x47, 0x50, 0x55, 0x18, 0x0d,
-	0x20, 0x01, 0x28, 0x09, 0x52, 0x07, 0x4d, 0x61, 0x69, 0x6e, 0x47, 0x50, 0x55, 0x12, 0x20, 0x0a,
-	0x0b, 0x54, 0x65, 0x6e, 0x73, 0x6f, 0x72, 0x53, 0x70, 0x6c, 0x69, 0x74, 0x18, 0x0e, 0x20, 0x01,
-	0x28, 0x09, 0x52, 0x0b, 0x54, 0x65, 0x6e, 0x73, 0x6f, 0x72, 0x53, 0x70, 0x6c, 0x69, 0x74, 0x12,
-	0x18, 0x0a, 0x07, 0x54, 0x68, 0x72, 0x65, 0x61, 0x64, 0x73, 0x18, 0x0f, 0x20, 0x01, 0x28, 0x05,
-	0x52, 0x07, 0x54, 0x68, 0x72, 0x65, 0x61, 0x64, 0x73, 0x12, 0x2c, 0x0a, 0x11, 0x4c, 0x69, 0x62,
-	0x72, 0x61, 0x72, 0x79, 0x53, 0x65, 0x61, 0x72, 0x63, 0x68, 0x50, 0x61, 0x74, 0x68, 0x18, 0x10,
-	0x20, 0x01, 0x28, 0x09, 0x52, 0x11, 0x4c, 0x69, 0x62, 0x72, 0x61, 0x72, 0x79, 0x53, 0x65, 0x61,
-	0x72, 0x63, 0x68, 0x50, 0x61, 0x74, 0x68, 0x12, 0x22, 0x0a, 0x0c, 0x52, 0x6f, 0x70, 0x65, 0x46,
-	0x72, 0x65, 0x71, 0x42, 0x61, 0x73, 0x65, 0x18, 0x11, 0x20, 0x01, 0x28, 0x02, 0x52, 0x0c, 0x52,
-	0x6f, 0x70, 0x65, 0x46, 0x72, 0x65, 0x71, 0x42, 0x61, 0x73, 0x65, 0x12, 0x24, 0x0a, 0x0d, 0x52,
-	0x6f, 0x70, 0x65, 0x46, 0x72, 0x65, 0x71, 0x53, 0x63, 0x61, 0x6c, 0x65, 0x18, 0x12, 0x20, 0x01,
-	0x28, 0x02, 0x52, 0x0d, 0x52, 0x6f, 0x70, 0x65, 0x46, 0x72, 0x65, 0x71, 0x53, 0x63, 0x61, 0x6c,
-	0x65, 0x12, 0x1e, 0x0a, 0x0a, 0x52, 0x4d, 0x53, 0x4e, 0x6f, 0x72, 0x6d, 0x45, 0x70, 0x73, 0x18,
-	0x13, 0x20, 0x01, 0x28, 0x02, 0x52, 0x0a, 0x52, 0x4d, 0x53, 0x4e, 0x6f, 0x72, 0x6d, 0x45, 0x70,
-	0x73, 0x12, 0x12, 0x0a, 0x04, 0x4e, 0x47, 0x51, 0x41, 0x18, 0x14, 0x20, 0x01, 0x28, 0x05, 0x52,
-	0x04, 0x4e, 0x47, 0x51, 0x41, 0x12, 0x1c, 0x0a, 0x09, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x46, 0x69,
-	0x6c, 0x65, 0x18, 0x15, 0x20, 0x01, 0x28, 0x09, 0x52, 0x09, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x46,
-	0x69, 0x6c, 0x65, 0x12, 0x16, 0x0a, 0x06, 0x44, 0x65, 0x76, 0x69, 0x63, 0x65, 0x18, 0x16, 0x20,
-	0x01, 0x28, 0x09, 0x52, 0x06, 0x44, 0x65, 0x76, 0x69, 0x63, 0x65, 0x12, 0x1c, 0x0a, 0x09, 0x55,
-	0x73, 0x65, 0x54, 0x72, 0x69, 0x74, 0x6f, 0x6e, 0x18, 0x17, 0x20, 0x01, 0x28, 0x08, 0x52, 0x09,
-	0x55, 0x73, 0x65, 0x54, 0x72, 0x69, 0x74, 0x6f, 0x6e, 0x12, 0x24, 0x0a, 0x0d, 0x4d, 0x6f, 0x64,
-	0x65, 0x6c, 0x42, 0x61, 0x73, 0x65, 0x4e, 0x61, 0x6d, 0x65, 0x18, 0x18, 0x20, 0x01, 0x28, 0x09,
-	0x52, 0x0d, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x42, 0x61, 0x73, 0x65, 0x4e, 0x61, 0x6d, 0x65, 0x12,
-	0x2a, 0x0a, 0x10, 0x55, 0x73, 0x65, 0x46, 0x61, 0x73, 0x74, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69,
-	0x7a, 0x65, 0x72, 0x18, 0x19, 0x20, 0x01, 0x28, 0x08, 0x52, 0x10, 0x55, 0x73, 0x65, 0x46, 0x61,
-	0x73, 0x74, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x65, 0x72, 0x12, 0x22, 0x0a, 0x0c, 0x50,
-	0x69, 0x70, 0x65, 0x6c, 0x69, 0x6e, 0x65, 0x54, 0x79, 0x70, 0x65, 0x18, 0x1a, 0x20, 0x01, 0x28,
-	0x09, 0x52, 0x0c, 0x50, 0x69, 0x70, 0x65, 0x6c, 0x69, 0x6e, 0x65, 0x54, 0x79, 0x70, 0x65, 0x12,
-	0x24, 0x0a, 0x0d, 0x53, 0x63, 0x68, 0x65, 0x64, 0x75, 0x6c, 0x65, 0x72, 0x54, 0x79, 0x70, 0x65,
-	0x18, 0x1b, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0d, 0x53, 0x63, 0x68, 0x65, 0x64, 0x75, 0x6c, 0x65,
-	0x72, 0x54, 0x79, 0x70, 0x65, 0x12, 0x12, 0x0a, 0x04, 0x43, 0x55, 0x44, 0x41, 0x18, 0x1c, 0x20,
-	0x01, 0x28, 0x08, 0x52, 0x04, 0x43, 0x55, 0x44, 0x41, 0x12, 0x1a, 0x0a, 0x08, 0x43, 0x46, 0x47,
-	0x53, 0x63, 0x61, 0x6c, 0x65, 0x18, 0x1d, 0x20, 0x01, 0x28, 0x02, 0x52, 0x08, 0x43, 0x46, 0x47,
-	0x53, 0x63, 0x61, 0x6c, 0x65, 0x12, 0x18, 0x0a, 0x07, 0x49, 0x4d, 0x47, 0x32, 0x49, 0x4d, 0x47,
-	0x18, 0x1e, 0x20, 0x01, 0x28, 0x08, 0x52, 0x07, 0x49, 0x4d, 0x47, 0x32, 0x49, 0x4d, 0x47, 0x12,
-	0x1c, 0x0a, 0x09, 0x43, 0x4c, 0x49, 0x50, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x18, 0x1f, 0x20, 0x01,
-	0x28, 0x09, 0x52, 0x09, 0x43, 0x4c, 0x49, 0x50, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x12, 0x24, 0x0a,
-	0x0d, 0x43, 0x4c, 0x49, 0x50, 0x53, 0x75, 0x62, 0x66, 0x6f, 0x6c, 0x64, 0x65, 0x72, 0x18, 0x20,
-	0x20, 0x01, 0x28, 0x09, 0x52, 0x0d, 0x43, 0x4c, 0x49, 0x50, 0x53, 0x75, 0x62, 0x66, 0x6f, 0x6c,
-	0x64, 0x65, 0x72, 0x12, 0x1a, 0x0a, 0x08, 0x43, 0x4c, 0x49, 0x50, 0x53, 0x6b, 0x69, 0x70, 0x18,
-	0x21, 0x20, 0x01, 0x28, 0x05, 0x52, 0x08, 0x43, 0x4c, 0x49, 0x50, 0x53, 0x6b, 0x69, 0x70, 0x12,
-	0x1c, 0x0a, 0x09, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x65, 0x72, 0x18, 0x22, 0x20, 0x01,
-	0x28, 0x09, 0x52, 0x09, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x65, 0x72, 0x12, 0x1a, 0x0a,
-	0x08, 0x4c, 0x6f, 0x72, 0x61, 0x42, 0x61, 0x73, 0x65, 0x18, 0x23, 0x20, 0x01, 0x28, 0x09, 0x52,
-	0x08, 0x4c, 0x6f, 0x72, 0x61, 0x42, 0x61, 0x73, 0x65, 0x12, 0x20, 0x0a, 0x0b, 0x4c, 0x6f, 0x72,
-	0x61, 0x41, 0x64, 0x61, 0x70, 0x74, 0x65, 0x72, 0x18, 0x24, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0b,
-	0x4c, 0x6f, 0x72, 0x61, 0x41, 0x64, 0x61, 0x70, 0x74, 0x65, 0x72, 0x12, 0x1c, 0x0a, 0x09, 0x4e,
-	0x6f, 0x4d, 0x75, 0x6c, 0x4d, 0x61, 0x74, 0x51, 0x18, 0x25, 0x20, 0x01, 0x28, 0x08, 0x52, 0x09,
-	0x4e, 0x6f, 0x4d, 0x75, 0x6c, 0x4d, 0x61, 0x74, 0x51, 0x12, 0x1e, 0x0a, 0x0a, 0x44, 0x72, 0x61,
-	0x66, 0x74, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x18, 0x27, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0a, 0x44,
-	0x72, 0x61, 0x66, 0x74, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x12, 0x1c, 0x0a, 0x09, 0x41, 0x75, 0x64,
-	0x69, 0x6f, 0x50, 0x61, 0x74, 0x68, 0x18, 0x26, 0x20, 0x01, 0x28, 0x09, 0x52, 0x09, 0x41, 0x75,
-	0x64, 0x69, 0x6f, 0x50, 0x61, 0x74, 0x68, 0x12, 0x22, 0x0a, 0x0c, 0x51, 0x75, 0x61, 0x6e, 0x74,
-	0x69, 0x7a, 0x61, 0x74, 0x69, 0x6f, 0x6e, 0x18, 0x28, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0c, 0x51,
-	0x75, 0x61, 0x6e, 0x74, 0x69, 0x7a, 0x61, 0x74, 0x69, 0x6f, 0x6e, 0x22, 0x3c, 0x0a, 0x06, 0x52,
-	0x65, 0x73, 0x75, 0x6c, 0x74, 0x12, 0x18, 0x0a, 0x07, 0x6d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65,
-	0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x07, 0x6d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x12,
-	0x18, 0x0a, 0x07, 0x73, 0x75, 0x63, 0x63, 0x65, 0x73, 0x73, 0x18, 0x02, 0x20, 0x01, 0x28, 0x08,
-	0x52, 0x07, 0x73, 0x75, 0x63, 0x63, 0x65, 0x73, 0x73, 0x22, 0x31, 0x0a, 0x0f, 0x45, 0x6d, 0x62,
-	0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x12, 0x1e, 0x0a, 0x0a,
-	0x65, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x73, 0x18, 0x01, 0x20, 0x03, 0x28, 0x02,
-	0x52, 0x0a, 0x65, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x73, 0x22, 0x5b, 0x0a, 0x11,
-	0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x52, 0x65, 0x71, 0x75, 0x65, 0x73,
-	0x74, 0x12, 0x10, 0x0a, 0x03, 0x64, 0x73, 0x74, 0x18, 0x02, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03,
-	0x64, 0x73, 0x74, 0x12, 0x1a, 0x0a, 0x08, 0x6c, 0x61, 0x6e, 0x67, 0x75, 0x61, 0x67, 0x65, 0x18,
-	0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x6c, 0x61, 0x6e, 0x67, 0x75, 0x61, 0x67, 0x65, 0x12,
-	0x18, 0x0a, 0x07, 0x74, 0x68, 0x72, 0x65, 0x61, 0x64, 0x73, 0x18, 0x04, 0x20, 0x01, 0x28, 0x0d,
-	0x52, 0x07, 0x74, 0x68, 0x72, 0x65, 0x61, 0x64, 0x73, 0x22, 0x5e, 0x0a, 0x10, 0x54, 0x72, 0x61,
-	0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x12, 0x36, 0x0a,
-	0x08, 0x73, 0x65, 0x67, 0x6d, 0x65, 0x6e, 0x74, 0x73, 0x18, 0x01, 0x20, 0x03, 0x28, 0x0b, 0x32,
+	0x06, 0x4e, 0x44, 0x72, 0x61, 0x66, 0x74, 0x12, 0x16, 0x0a, 0x06, 0x49, 0x6d, 0x61, 0x67, 0x65,
+	0x73, 0x18, 0x2a, 0x20, 0x03, 0x28, 0x09, 0x52, 0x06, 0x49, 0x6d, 0x61, 0x67, 0x65, 0x73, 0x22,
+	0x21, 0x0a, 0x05, 0x52, 0x65, 0x70, 0x6c, 0x79, 0x12, 0x18, 0x0a, 0x07, 0x6d, 0x65, 0x73, 0x73,
+	0x61, 0x67, 0x65, 0x18, 0x01, 0x20, 0x01, 0x28, 0x0c, 0x52, 0x07, 0x6d, 0x65, 0x73, 0x73, 0x61,
+	0x67, 0x65, 0x22, 0xac, 0x0b, 0x0a, 0x0c, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x4f, 0x70, 0x74, 0x69,
+	0x6f, 0x6e, 0x73, 0x12, 0x14, 0x0a, 0x05, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x18, 0x01, 0x20, 0x01,
+	0x28, 0x09, 0x52, 0x05, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x12, 0x20, 0x0a, 0x0b, 0x43, 0x6f, 0x6e,
+	0x74, 0x65, 0x78, 0x74, 0x53, 0x69, 0x7a, 0x65, 0x18, 0x02, 0x20, 0x01, 0x28, 0x05, 0x52, 0x0b,
+	0x43, 0x6f, 0x6e, 0x74, 0x65, 0x78, 0x74, 0x53, 0x69, 0x7a, 0x65, 0x12, 0x12, 0x0a, 0x04, 0x53,
+	0x65, 0x65, 0x64, 0x18, 0x03, 0x20, 0x01, 0x28, 0x05, 0x52, 0x04, 0x53, 0x65, 0x65, 0x64, 0x12,
+	0x16, 0x0a, 0x06, 0x4e, 0x42, 0x61, 0x74, 0x63, 0x68, 0x18, 0x04, 0x20, 0x01, 0x28, 0x05, 0x52,
+	0x06, 0x4e, 0x42, 0x61, 0x74, 0x63, 0x68, 0x12, 0x1c, 0x0a, 0x09, 0x46, 0x31, 0x36, 0x4d, 0x65,
+	0x6d, 0x6f, 0x72, 0x79, 0x18, 0x05, 0x20, 0x01, 0x28, 0x08, 0x52, 0x09, 0x46, 0x31, 0x36, 0x4d,
+	0x65, 0x6d, 0x6f, 0x72, 0x79, 0x12, 0x14, 0x0a, 0x05, 0x4d, 0x4c, 0x6f, 0x63, 0x6b, 0x18, 0x06,
+	0x20, 0x01, 0x28, 0x08, 0x52, 0x05, 0x4d, 0x4c, 0x6f, 0x63, 0x6b, 0x12, 0x12, 0x0a, 0x04, 0x4d,
+	0x4d, 0x61, 0x70, 0x18, 0x07, 0x20, 0x01, 0x28, 0x08, 0x52, 0x04, 0x4d, 0x4d, 0x61, 0x70, 0x12,
+	0x1c, 0x0a, 0x09, 0x56, 0x6f, 0x63, 0x61, 0x62, 0x4f, 0x6e, 0x6c, 0x79, 0x18, 0x08, 0x20, 0x01,
+	0x28, 0x08, 0x52, 0x09, 0x56, 0x6f, 0x63, 0x61, 0x62, 0x4f, 0x6e, 0x6c, 0x79, 0x12, 0x18, 0x0a,
+	0x07, 0x4c, 0x6f, 0x77, 0x56, 0x52, 0x41, 0x4d, 0x18, 0x09, 0x20, 0x01, 0x28, 0x08, 0x52, 0x07,
+	0x4c, 0x6f, 0x77, 0x56, 0x52, 0x41, 0x4d, 0x12, 0x1e, 0x0a, 0x0a, 0x45, 0x6d, 0x62, 0x65, 0x64,
+	0x64, 0x69, 0x6e, 0x67, 0x73, 0x18, 0x0a, 0x20, 0x01, 0x28, 0x08, 0x52, 0x0a, 0x45, 0x6d, 0x62,
+	0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x73, 0x12, 0x12, 0x0a, 0x04, 0x4e, 0x55, 0x4d, 0x41, 0x18,
+	0x0b, 0x20, 0x01, 0x28, 0x08, 0x52, 0x04, 0x4e, 0x55, 0x4d, 0x41, 0x12, 0x1e, 0x0a, 0x0a, 0x4e,
+	0x47, 0x50, 0x55, 0x4c, 0x61, 0x79, 0x65, 0x72, 0x73, 0x18, 0x0c, 0x20, 0x01, 0x28, 0x05, 0x52,
+	0x0a, 0x4e, 0x47, 0x50, 0x55, 0x4c, 0x61, 0x79, 0x65, 0x72, 0x73, 0x12, 0x18, 0x0a, 0x07, 0x4d,
+	0x61, 0x69, 0x6e, 0x47, 0x50, 0x55, 0x18, 0x0d, 0x20, 0x01, 0x28, 0x09, 0x52, 0x07, 0x4d, 0x61,
+	0x69, 0x6e, 0x47, 0x50, 0x55, 0x12, 0x20, 0x0a, 0x0b, 0x54, 0x65, 0x6e, 0x73, 0x6f, 0x72, 0x53,
+	0x70, 0x6c, 0x69, 0x74, 0x18, 0x0e, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0b, 0x54, 0x65, 0x6e, 0x73,
+	0x6f, 0x72, 0x53, 0x70, 0x6c, 0x69, 0x74, 0x12, 0x18, 0x0a, 0x07, 0x54, 0x68, 0x72, 0x65, 0x61,
+	0x64, 0x73, 0x18, 0x0f, 0x20, 0x01, 0x28, 0x05, 0x52, 0x07, 0x54, 0x68, 0x72, 0x65, 0x61, 0x64,
+	0x73, 0x12, 0x2c, 0x0a, 0x11, 0x4c, 0x69, 0x62, 0x72, 0x61, 0x72, 0x79, 0x53, 0x65, 0x61, 0x72,
+	0x63, 0x68, 0x50, 0x61, 0x74, 0x68, 0x18, 0x10, 0x20, 0x01, 0x28, 0x09, 0x52, 0x11, 0x4c, 0x69,
+	0x62, 0x72, 0x61, 0x72, 0x79, 0x53, 0x65, 0x61, 0x72, 0x63, 0x68, 0x50, 0x61, 0x74, 0x68, 0x12,
+	0x22, 0x0a, 0x0c, 0x52, 0x6f, 0x70, 0x65, 0x46, 0x72, 0x65, 0x71, 0x42, 0x61, 0x73, 0x65, 0x18,
+	0x11, 0x20, 0x01, 0x28, 0x02, 0x52, 0x0c, 0x52, 0x6f, 0x70, 0x65, 0x46, 0x72, 0x65, 0x71, 0x42,
+	0x61, 0x73, 0x65, 0x12, 0x24, 0x0a, 0x0d, 0x52, 0x6f, 0x70, 0x65, 0x46, 0x72, 0x65, 0x71, 0x53,
+	0x63, 0x61, 0x6c, 0x65, 0x18, 0x12, 0x20, 0x01, 0x28, 0x02, 0x52, 0x0d, 0x52, 0x6f, 0x70, 0x65,
+	0x46, 0x72, 0x65, 0x71, 0x53, 0x63, 0x61, 0x6c, 0x65, 0x12, 0x1e, 0x0a, 0x0a, 0x52, 0x4d, 0x53,
+	0x4e, 0x6f, 0x72, 0x6d, 0x45, 0x70, 0x73, 0x18, 0x13, 0x20, 0x01, 0x28, 0x02, 0x52, 0x0a, 0x52,
+	0x4d, 0x53, 0x4e, 0x6f, 0x72, 0x6d, 0x45, 0x70, 0x73, 0x12, 0x12, 0x0a, 0x04, 0x4e, 0x47, 0x51,
+	0x41, 0x18, 0x14, 0x20, 0x01, 0x28, 0x05, 0x52, 0x04, 0x4e, 0x47, 0x51, 0x41, 0x12, 0x1c, 0x0a,
+	0x09, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x46, 0x69, 0x6c, 0x65, 0x18, 0x15, 0x20, 0x01, 0x28, 0x09,
+	0x52, 0x09, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x46, 0x69, 0x6c, 0x65, 0x12, 0x16, 0x0a, 0x06, 0x44,
+	0x65, 0x76, 0x69, 0x63, 0x65, 0x18, 0x16, 0x20, 0x01, 0x28, 0x09, 0x52, 0x06, 0x44, 0x65, 0x76,
+	0x69, 0x63, 0x65, 0x12, 0x1c, 0x0a, 0x09, 0x55, 0x73, 0x65, 0x54, 0x72, 0x69, 0x74, 0x6f, 0x6e,
+	0x18, 0x17, 0x20, 0x01, 0x28, 0x08, 0x52, 0x09, 0x55, 0x73, 0x65, 0x54, 0x72, 0x69, 0x74, 0x6f,
+	0x6e, 0x12, 0x24, 0x0a, 0x0d, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x42, 0x61, 0x73, 0x65, 0x4e, 0x61,
+	0x6d, 0x65, 0x18, 0x18, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0d, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x42,
+	0x61, 0x73, 0x65, 0x4e, 0x61, 0x6d, 0x65, 0x12, 0x2a, 0x0a, 0x10, 0x55, 0x73, 0x65, 0x46, 0x61,
+	0x73, 0x74, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x65, 0x72, 0x18, 0x19, 0x20, 0x01, 0x28,
+	0x08, 0x52, 0x10, 0x55, 0x73, 0x65, 0x46, 0x61, 0x73, 0x74, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69,
+	0x7a, 0x65, 0x72, 0x12, 0x22, 0x0a, 0x0c, 0x50, 0x69, 0x70, 0x65, 0x6c, 0x69, 0x6e, 0x65, 0x54,
+	0x79, 0x70, 0x65, 0x18, 0x1a, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0c, 0x50, 0x69, 0x70, 0x65, 0x6c,
+	0x69, 0x6e, 0x65, 0x54, 0x79, 0x70, 0x65, 0x12, 0x24, 0x0a, 0x0d, 0x53, 0x63, 0x68, 0x65, 0x64,
+	0x75, 0x6c, 0x65, 0x72, 0x54, 0x79, 0x70, 0x65, 0x18, 0x1b, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0d,
+	0x53, 0x63, 0x68, 0x65, 0x64, 0x75, 0x6c, 0x65, 0x72, 0x54, 0x79, 0x70, 0x65, 0x12, 0x12, 0x0a,
+	0x04, 0x43, 0x55, 0x44, 0x41, 0x18, 0x1c, 0x20, 0x01, 0x28, 0x08, 0x52, 0x04, 0x43, 0x55, 0x44,
+	0x41, 0x12, 0x1a, 0x0a, 0x08, 0x43, 0x46, 0x47, 0x53, 0x63, 0x61, 0x6c, 0x65, 0x18, 0x1d, 0x20,
+	0x01, 0x28, 0x02, 0x52, 0x08, 0x43, 0x46, 0x47, 0x53, 0x63, 0x61, 0x6c, 0x65, 0x12, 0x18, 0x0a,
+	0x07, 0x49, 0x4d, 0x47, 0x32, 0x49, 0x4d, 0x47, 0x18, 0x1e, 0x20, 0x01, 0x28, 0x08, 0x52, 0x07,
+	0x49, 0x4d, 0x47, 0x32, 0x49, 0x4d, 0x47, 0x12, 0x1c, 0x0a, 0x09, 0x43, 0x4c, 0x49, 0x50, 0x4d,
+	0x6f, 0x64, 0x65, 0x6c, 0x18, 0x1f, 0x20, 0x01, 0x28, 0x09, 0x52, 0x09, 0x43, 0x4c, 0x49, 0x50,
+	0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x12, 0x24, 0x0a, 0x0d, 0x43, 0x4c, 0x49, 0x50, 0x53, 0x75, 0x62,
+	0x66, 0x6f, 0x6c, 0x64, 0x65, 0x72, 0x18, 0x20, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0d, 0x43, 0x4c,
+	0x49, 0x50, 0x53, 0x75, 0x62, 0x66, 0x6f, 0x6c, 0x64, 0x65, 0x72, 0x12, 0x1a, 0x0a, 0x08, 0x43,
+	0x4c, 0x49, 0x50, 0x53, 0x6b, 0x69, 0x70, 0x18, 0x21, 0x20, 0x01, 0x28, 0x05, 0x52, 0x08, 0x43,
+	0x4c, 0x49, 0x50, 0x53, 0x6b, 0x69, 0x70, 0x12, 0x1c, 0x0a, 0x09, 0x54, 0x6f, 0x6b, 0x65, 0x6e,
+	0x69, 0x7a, 0x65, 0x72, 0x18, 0x22, 0x20, 0x01, 0x28, 0x09, 0x52, 0x09, 0x54, 0x6f, 0x6b, 0x65,
+	0x6e, 0x69, 0x7a, 0x65, 0x72, 0x12, 0x1a, 0x0a, 0x08, 0x4c, 0x6f, 0x72, 0x61, 0x42, 0x61, 0x73,
+	0x65, 0x18, 0x23, 0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x4c, 0x6f, 0x72, 0x61, 0x42, 0x61, 0x73,
+	0x65, 0x12, 0x20, 0x0a, 0x0b, 0x4c, 0x6f, 0x72, 0x61, 0x41, 0x64, 0x61, 0x70, 0x74, 0x65, 0x72,
+	0x18, 0x24, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0b, 0x4c, 0x6f, 0x72, 0x61, 0x41, 0x64, 0x61, 0x70,
+	0x74, 0x65, 0x72, 0x12, 0x1c, 0x0a, 0x09, 0x4c, 0x6f, 0x72, 0x61, 0x53, 0x63, 0x61, 0x6c, 0x65,
+	0x18, 0x2a, 0x20, 0x01, 0x28, 0x02, 0x52, 0x09, 0x4c, 0x6f, 0x72, 0x61, 0x53, 0x63, 0x61, 0x6c,
+	0x65, 0x12, 0x1c, 0x0a, 0x09, 0x4e, 0x6f, 0x4d, 0x75, 0x6c, 0x4d, 0x61, 0x74, 0x51, 0x18, 0x25,
+	0x20, 0x01, 0x28, 0x08, 0x52, 0x09, 0x4e, 0x6f, 0x4d, 0x75, 0x6c, 0x4d, 0x61, 0x74, 0x51, 0x12,
+	0x1e, 0x0a, 0x0a, 0x44, 0x72, 0x61, 0x66, 0x74, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x18, 0x27, 0x20,
+	0x01, 0x28, 0x09, 0x52, 0x0a, 0x44, 0x72, 0x61, 0x66, 0x74, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x12,
+	0x1c, 0x0a, 0x09, 0x41, 0x75, 0x64, 0x69, 0x6f, 0x50, 0x61, 0x74, 0x68, 0x18, 0x26, 0x20, 0x01,
+	0x28, 0x09, 0x52, 0x09, 0x41, 0x75, 0x64, 0x69, 0x6f, 0x50, 0x61, 0x74, 0x68, 0x12, 0x22, 0x0a,
+	0x0c, 0x51, 0x75, 0x61, 0x6e, 0x74, 0x69, 0x7a, 0x61, 0x74, 0x69, 0x6f, 0x6e, 0x18, 0x28, 0x20,
+	0x01, 0x28, 0x09, 0x52, 0x0c, 0x51, 0x75, 0x61, 0x6e, 0x74, 0x69, 0x7a, 0x61, 0x74, 0x69, 0x6f,
+	0x6e, 0x12, 0x16, 0x0a, 0x06, 0x4d, 0x4d, 0x50, 0x72, 0x6f, 0x6a, 0x18, 0x29, 0x20, 0x01, 0x28,
+	0x09, 0x52, 0x06, 0x4d, 0x4d, 0x50, 0x72, 0x6f, 0x6a, 0x12, 0x20, 0x0a, 0x0b, 0x52, 0x6f, 0x70,
+	0x65, 0x53, 0x63, 0x61, 0x6c, 0x69, 0x6e, 0x67, 0x18, 0x2b, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0b,
+	0x52, 0x6f, 0x70, 0x65, 0x53, 0x63, 0x61, 0x6c, 0x69, 0x6e, 0x67, 0x12, 0x24, 0x0a, 0x0d, 0x59,
+	0x61, 0x72, 0x6e, 0x45, 0x78, 0x74, 0x46, 0x61, 0x63, 0x74, 0x6f, 0x72, 0x18, 0x2c, 0x20, 0x01,
+	0x28, 0x02, 0x52, 0x0d, 0x59, 0x61, 0x72, 0x6e, 0x45, 0x78, 0x74, 0x46, 0x61, 0x63, 0x74, 0x6f,
+	0x72, 0x12, 0x26, 0x0a, 0x0e, 0x59, 0x61, 0x72, 0x6e, 0x41, 0x74, 0x74, 0x6e, 0x46, 0x61, 0x63,
+	0x74, 0x6f, 0x72, 0x18, 0x2d, 0x20, 0x01, 0x28, 0x02, 0x52, 0x0e, 0x59, 0x61, 0x72, 0x6e, 0x41,
+	0x74, 0x74, 0x6e, 0x46, 0x61, 0x63, 0x74, 0x6f, 0x72, 0x12, 0x22, 0x0a, 0x0c, 0x59, 0x61, 0x72,
+	0x6e, 0x42, 0x65, 0x74, 0x61, 0x46, 0x61, 0x73, 0x74, 0x18, 0x2e, 0x20, 0x01, 0x28, 0x02, 0x52,
+	0x0c, 0x59, 0x61, 0x72, 0x6e, 0x42, 0x65, 0x74, 0x61, 0x46, 0x61, 0x73, 0x74, 0x12, 0x22, 0x0a,
+	0x0c, 0x59, 0x61, 0x72, 0x6e, 0x42, 0x65, 0x74, 0x61, 0x53, 0x6c, 0x6f, 0x77, 0x18, 0x2f, 0x20,
+	0x01, 0x28, 0x02, 0x52, 0x0c, 0x59, 0x61, 0x72, 0x6e, 0x42, 0x65, 0x74, 0x61, 0x53, 0x6c, 0x6f,
+	0x77, 0x22, 0x3c, 0x0a, 0x06, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x12, 0x18, 0x0a, 0x07, 0x6d,
+	0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x07, 0x6d, 0x65,
+	0x73, 0x73, 0x61, 0x67, 0x65, 0x12, 0x18, 0x0a, 0x07, 0x73, 0x75, 0x63, 0x63, 0x65, 0x73, 0x73,
+	0x18, 0x02, 0x20, 0x01, 0x28, 0x08, 0x52, 0x07, 0x73, 0x75, 0x63, 0x63, 0x65, 0x73, 0x73, 0x22,
+	0x31, 0x0a, 0x0f, 0x45, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x52, 0x65, 0x73, 0x75,
+	0x6c, 0x74, 0x12, 0x1e, 0x0a, 0x0a, 0x65, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x73,
+	0x18, 0x01, 0x20, 0x03, 0x28, 0x02, 0x52, 0x0a, 0x65, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e,
+	0x67, 0x73, 0x22, 0x5b, 0x0a, 0x11, 0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74,
+	0x52, 0x65, 0x71, 0x75, 0x65, 0x73, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x64, 0x73, 0x74, 0x18, 0x02,
+	0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x64, 0x73, 0x74, 0x12, 0x1a, 0x0a, 0x08, 0x6c, 0x61, 0x6e,
+	0x67, 0x75, 0x61, 0x67, 0x65, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x6c, 0x61, 0x6e,
+	0x67, 0x75, 0x61, 0x67, 0x65, 0x12, 0x18, 0x0a, 0x07, 0x74, 0x68, 0x72, 0x65, 0x61, 0x64, 0x73,
+	0x18, 0x04, 0x20, 0x01, 0x28, 0x0d, 0x52, 0x07, 0x74, 0x68, 0x72, 0x65, 0x61, 0x64, 0x73, 0x22,
+	0x5e, 0x0a, 0x10, 0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x52, 0x65, 0x73,
+	0x75, 0x6c, 0x74, 0x12, 0x36, 0x0a, 0x08, 0x73, 0x65, 0x67, 0x6d, 0x65, 0x6e, 0x74, 0x73, 0x18,
+	0x01, 0x20, 0x03, 0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e,
+	0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x53, 0x65, 0x67, 0x6d, 0x65, 0x6e,
+	0x74, 0x52, 0x08, 0x73, 0x65, 0x67, 0x6d, 0x65, 0x6e, 0x74, 0x73, 0x12, 0x12, 0x0a, 0x04, 0x74,
+	0x65, 0x78, 0x74, 0x18, 0x02, 0x20, 0x01, 0x28, 0x09, 0x52, 0x04, 0x74, 0x65, 0x78, 0x74, 0x22,
+	0x77, 0x0a, 0x11, 0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x53, 0x65, 0x67,
+	0x6d, 0x65, 0x6e, 0x74, 0x12, 0x0e, 0x0a, 0x02, 0x69, 0x64, 0x18, 0x01, 0x20, 0x01, 0x28, 0x05,
+	0x52, 0x02, 0x69, 0x64, 0x12, 0x14, 0x0a, 0x05, 0x73, 0x74, 0x61, 0x72, 0x74, 0x18, 0x02, 0x20,
+	0x01, 0x28, 0x03, 0x52, 0x05, 0x73, 0x74, 0x61, 0x72, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x65, 0x6e,
+	0x64, 0x18, 0x03, 0x20, 0x01, 0x28, 0x03, 0x52, 0x03, 0x65, 0x6e, 0x64, 0x12, 0x12, 0x0a, 0x04,
+	0x74, 0x65, 0x78, 0x74, 0x18, 0x04, 0x20, 0x01, 0x28, 0x09, 0x52, 0x04, 0x74, 0x65, 0x78, 0x74,
+	0x12, 0x16, 0x0a, 0x06, 0x74, 0x6f, 0x6b, 0x65, 0x6e, 0x73, 0x18, 0x05, 0x20, 0x03, 0x28, 0x05,
+	0x52, 0x06, 0x74, 0x6f, 0x6b, 0x65, 0x6e, 0x73, 0x22, 0xbe, 0x02, 0x0a, 0x14, 0x47, 0x65, 0x6e,
+	0x65, 0x72, 0x61, 0x74, 0x65, 0x49, 0x6d, 0x61, 0x67, 0x65, 0x52, 0x65, 0x71, 0x75, 0x65, 0x73,
+	0x74, 0x12, 0x16, 0x0a, 0x06, 0x68, 0x65, 0x69, 0x67, 0x68, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28,
+	0x05, 0x52, 0x06, 0x68, 0x65, 0x69, 0x67, 0x68, 0x74, 0x12, 0x14, 0x0a, 0x05, 0x77, 0x69, 0x64,
+	0x74, 0x68, 0x18, 0x02, 0x20, 0x01, 0x28, 0x05, 0x52, 0x05, 0x77, 0x69, 0x64, 0x74, 0x68, 0x12,
+	0x12, 0x0a, 0x04, 0x6d, 0x6f, 0x64, 0x65, 0x18, 0x03, 0x20, 0x01, 0x28, 0x05, 0x52, 0x04, 0x6d,
+	0x6f, 0x64, 0x65, 0x12, 0x12, 0x0a, 0x04, 0x73, 0x74, 0x65, 0x70, 0x18, 0x04, 0x20, 0x01, 0x28,
+	0x05, 0x52, 0x04, 0x73, 0x74, 0x65, 0x70, 0x12, 0x12, 0x0a, 0x04, 0x73, 0x65, 0x65, 0x64, 0x18,
+	0x05, 0x20, 0x01, 0x28, 0x05, 0x52, 0x04, 0x73, 0x65, 0x65, 0x64, 0x12, 0x27, 0x0a, 0x0f, 0x70,
+	0x6f, 0x73, 0x69, 0x74, 0x69, 0x76, 0x65, 0x5f, 0x70, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x18, 0x06,
+	0x20, 0x01, 0x28, 0x09, 0x52, 0x0e, 0x70, 0x6f, 0x73, 0x69, 0x74, 0x69, 0x76, 0x65, 0x50, 0x72,
+	0x6f, 0x6d, 0x70, 0x74, 0x12, 0x27, 0x0a, 0x0f, 0x6e, 0x65, 0x67, 0x61, 0x74, 0x69, 0x76, 0x65,
+	0x5f, 0x70, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x18, 0x07, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0e, 0x6e,
+	0x65, 0x67, 0x61, 0x74, 0x69, 0x76, 0x65, 0x50, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x12, 0x10, 0x0a,
+	0x03, 0x64, 0x73, 0x74, 0x18, 0x08, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x64, 0x73, 0x74, 0x12,
+	0x10, 0x0a, 0x03, 0x73, 0x72, 0x63, 0x18, 0x09, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x73, 0x72,
+	0x63, 0x12, 0x2a, 0x0a, 0x10, 0x45, 0x6e, 0x61, 0x62, 0x6c, 0x65, 0x50, 0x61, 0x72, 0x61, 0x6d,
+	0x65, 0x74, 0x65, 0x72, 0x73, 0x18, 0x0a, 0x20, 0x01, 0x28, 0x09, 0x52, 0x10, 0x45, 0x6e, 0x61,
+	0x62, 0x6c, 0x65, 0x50, 0x61, 0x72, 0x61, 0x6d, 0x65, 0x74, 0x65, 0x72, 0x73, 0x12, 0x1a, 0x0a,
+	0x08, 0x43, 0x4c, 0x49, 0x50, 0x53, 0x6b, 0x69, 0x70, 0x18, 0x0b, 0x20, 0x01, 0x28, 0x05, 0x52,
+	0x08, 0x43, 0x4c, 0x49, 0x50, 0x53, 0x6b, 0x69, 0x70, 0x22, 0x48, 0x0a, 0x0a, 0x54, 0x54, 0x53,
+	0x52, 0x65, 0x71, 0x75, 0x65, 0x73, 0x74, 0x12, 0x12, 0x0a, 0x04, 0x74, 0x65, 0x78, 0x74, 0x18,
+	0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x04, 0x74, 0x65, 0x78, 0x74, 0x12, 0x14, 0x0a, 0x05, 0x6d,
+	0x6f, 0x64, 0x65, 0x6c, 0x18, 0x02, 0x20, 0x01, 0x28, 0x09, 0x52, 0x05, 0x6d, 0x6f, 0x64, 0x65,
+	0x6c, 0x12, 0x10, 0x0a, 0x03, 0x64, 0x73, 0x74, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03,
+	0x64, 0x73, 0x74, 0x22, 0x46, 0x0a, 0x14, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x61, 0x74,
+	0x69, 0x6f, 0x6e, 0x52, 0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73, 0x65, 0x12, 0x16, 0x0a, 0x06, 0x6c,
+	0x65, 0x6e, 0x67, 0x74, 0x68, 0x18, 0x01, 0x20, 0x01, 0x28, 0x05, 0x52, 0x06, 0x6c, 0x65, 0x6e,
+	0x67, 0x74, 0x68, 0x12, 0x16, 0x0a, 0x06, 0x74, 0x6f, 0x6b, 0x65, 0x6e, 0x73, 0x18, 0x02, 0x20,
+	0x03, 0x28, 0x05, 0x52, 0x06, 0x74, 0x6f, 0x6b, 0x65, 0x6e, 0x73, 0x22, 0xac, 0x01, 0x0a, 0x0f,
+	0x4d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x55, 0x73, 0x61, 0x67, 0x65, 0x44, 0x61, 0x74, 0x61, 0x12,
+	0x14, 0x0a, 0x05, 0x74, 0x6f, 0x74, 0x61, 0x6c, 0x18, 0x01, 0x20, 0x01, 0x28, 0x04, 0x52, 0x05,
+	0x74, 0x6f, 0x74, 0x61, 0x6c, 0x12, 0x45, 0x0a, 0x09, 0x62, 0x72, 0x65, 0x61, 0x6b, 0x64, 0x6f,
+	0x77, 0x6e, 0x18, 0x02, 0x20, 0x03, 0x28, 0x0b, 0x32, 0x27, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65,
+	0x6e, 0x64, 0x2e, 0x4d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x55, 0x73, 0x61, 0x67, 0x65, 0x44, 0x61,
+	0x74, 0x61, 0x2e, 0x42, 0x72, 0x65, 0x61, 0x6b, 0x64, 0x6f, 0x77, 0x6e, 0x45, 0x6e, 0x74, 0x72,
+	0x79, 0x52, 0x09, 0x62, 0x72, 0x65, 0x61, 0x6b, 0x64, 0x6f, 0x77, 0x6e, 0x1a, 0x3c, 0x0a, 0x0e,
+	0x42, 0x72, 0x65, 0x61, 0x6b, 0x64, 0x6f, 0x77, 0x6e, 0x45, 0x6e, 0x74, 0x72, 0x79, 0x12, 0x10,
+	0x0a, 0x03, 0x6b, 0x65, 0x79, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x6b, 0x65, 0x79,
+	0x12, 0x14, 0x0a, 0x05, 0x76, 0x61, 0x6c, 0x75, 0x65, 0x18, 0x02, 0x20, 0x01, 0x28, 0x04, 0x52,
+	0x05, 0x76, 0x61, 0x6c, 0x75, 0x65, 0x3a, 0x02, 0x38, 0x01, 0x22, 0xbc, 0x01, 0x0a, 0x0e, 0x53,
+	0x74, 0x61, 0x74, 0x75, 0x73, 0x52, 0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73, 0x65, 0x12, 0x33, 0x0a,
+	0x05, 0x73, 0x74, 0x61, 0x74, 0x65, 0x18, 0x01, 0x20, 0x01, 0x28, 0x0e, 0x32, 0x1d, 0x2e, 0x62,
+	0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x53, 0x74, 0x61, 0x74, 0x75, 0x73, 0x52, 0x65, 0x73,
+	0x70, 0x6f, 0x6e, 0x73, 0x65, 0x2e, 0x53, 0x74, 0x61, 0x74, 0x65, 0x52, 0x05, 0x73, 0x74, 0x61,
+	0x74, 0x65, 0x12, 0x30, 0x0a, 0x06, 0x6d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x18, 0x02, 0x20, 0x01,
+	0x28, 0x0b, 0x32, 0x18, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x4d, 0x65, 0x6d,
+	0x6f, 0x72, 0x79, 0x55, 0x73, 0x61, 0x67, 0x65, 0x44, 0x61, 0x74, 0x61, 0x52, 0x06, 0x6d, 0x65,
+	0x6d, 0x6f, 0x72, 0x79, 0x22, 0x43, 0x0a, 0x05, 0x53, 0x74, 0x61, 0x74, 0x65, 0x12, 0x11, 0x0a,
+	0x0d, 0x55, 0x4e, 0x49, 0x4e, 0x49, 0x54, 0x49, 0x41, 0x4c, 0x49, 0x5a, 0x45, 0x44, 0x10, 0x00,
+	0x12, 0x08, 0x0a, 0x04, 0x42, 0x55, 0x53, 0x59, 0x10, 0x01, 0x12, 0x09, 0x0a, 0x05, 0x52, 0x45,
+	0x41, 0x44, 0x59, 0x10, 0x02, 0x12, 0x12, 0x0a, 0x05, 0x45, 0x52, 0x52, 0x4f, 0x52, 0x10, 0xff,
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x01, 0x32, 0xf4, 0x04, 0x0a, 0x07, 0x42, 0x61,
+	0x63, 0x6b, 0x65, 0x6e, 0x64, 0x12, 0x32, 0x0a, 0x06, 0x48, 0x65, 0x61, 0x6c, 0x74, 0x68, 0x12,
+	0x16, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x48, 0x65, 0x61, 0x6c, 0x74, 0x68,
+	0x4d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x1a, 0x0e, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e,
+	0x64, 0x2e, 0x52, 0x65, 0x70, 0x6c, 0x79, 0x22, 0x00, 0x12, 0x34, 0x0a, 0x07, 0x50, 0x72, 0x65,
+	0x64, 0x69, 0x63, 0x74, 0x12, 0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x50,
+	0x72, 0x65, 0x64, 0x69, 0x63, 0x74, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x1a, 0x0e, 0x2e,
+	0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65, 0x70, 0x6c, 0x79, 0x22, 0x00, 0x12,
+	0x35, 0x0a, 0x09, 0x4c, 0x6f, 0x61, 0x64, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x12, 0x15, 0x2e, 0x62,
+	0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x4f, 0x70, 0x74, 0x69,
+	0x6f, 0x6e, 0x73, 0x1a, 0x0f, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65,
+	0x73, 0x75, 0x6c, 0x74, 0x22, 0x00, 0x12, 0x3c, 0x0a, 0x0d, 0x50, 0x72, 0x65, 0x64, 0x69, 0x63,
+	0x74, 0x53, 0x74, 0x72, 0x65, 0x61, 0x6d, 0x12, 0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e,
+	0x64, 0x2e, 0x50, 0x72, 0x65, 0x64, 0x69, 0x63, 0x74, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73,
+	0x1a, 0x0e, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65, 0x70, 0x6c, 0x79,
+	0x22, 0x00, 0x30, 0x01, 0x12, 0x40, 0x0a, 0x09, 0x45, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e,
+	0x67, 0x12, 0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x50, 0x72, 0x65, 0x64,
+	0x69, 0x63, 0x74, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x1a, 0x18, 0x2e, 0x62, 0x61, 0x63,
+	0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x45, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x52, 0x65,
+	0x73, 0x75, 0x6c, 0x74, 0x22, 0x00, 0x12, 0x41, 0x0a, 0x0d, 0x47, 0x65, 0x6e, 0x65, 0x72, 0x61,
+	0x74, 0x65, 0x49, 0x6d, 0x61, 0x67, 0x65, 0x12, 0x1d, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e,
+	0x64, 0x2e, 0x47, 0x65, 0x6e, 0x65, 0x72, 0x61, 0x74, 0x65, 0x49, 0x6d, 0x61, 0x67, 0x65, 0x52,
+	0x65, 0x71, 0x75, 0x65, 0x73, 0x74, 0x1a, 0x0f, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64,
+	0x2e, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x22, 0x00, 0x12, 0x4d, 0x0a, 0x12, 0x41, 0x75, 0x64,
+	0x69, 0x6f, 0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x12,
 	0x1a, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x54, 0x72, 0x61, 0x6e, 0x73, 0x63,
-	0x72, 0x69, 0x70, 0x74, 0x53, 0x65, 0x67, 0x6d, 0x65, 0x6e, 0x74, 0x52, 0x08, 0x73, 0x65, 0x67,
-	0x6d, 0x65, 0x6e, 0x74, 0x73, 0x12, 0x12, 0x0a, 0x04, 0x74, 0x65, 0x78, 0x74, 0x18, 0x02, 0x20,
-	0x01, 0x28, 0x09, 0x52, 0x04, 0x74, 0x65, 0x78, 0x74, 0x22, 0x77, 0x0a, 0x11, 0x54, 0x72, 0x61,
-	0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x53, 0x65, 0x67, 0x6d, 0x65, 0x6e, 0x74, 0x12, 0x0e,
-	0x0a, 0x02, 0x69, 0x64, 0x18, 0x01, 0x20, 0x01, 0x28, 0x05, 0x52, 0x02, 0x69, 0x64, 0x12, 0x14,
-	0x0a, 0x05, 0x73, 0x74, 0x61, 0x72, 0x74, 0x18, 0x02, 0x20, 0x01, 0x28, 0x03, 0x52, 0x05, 0x73,
-	0x74, 0x61, 0x72, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x65, 0x6e, 0x64, 0x18, 0x03, 0x20, 0x01, 0x28,
-	0x03, 0x52, 0x03, 0x65, 0x6e, 0x64, 0x12, 0x12, 0x0a, 0x04, 0x74, 0x65, 0x78, 0x74, 0x18, 0x04,
-	0x20, 0x01, 0x28, 0x09, 0x52, 0x04, 0x74, 0x65, 0x78, 0x74, 0x12, 0x16, 0x0a, 0x06, 0x74, 0x6f,
-	0x6b, 0x65, 0x6e, 0x73, 0x18, 0x05, 0x20, 0x03, 0x28, 0x05, 0x52, 0x06, 0x74, 0x6f, 0x6b, 0x65,
-	0x6e, 0x73, 0x22, 0xbe, 0x02, 0x0a, 0x14, 0x47, 0x65, 0x6e, 0x65, 0x72, 0x61, 0x74, 0x65, 0x49,
-	0x6d, 0x61, 0x67, 0x65, 0x52, 0x65, 0x71, 0x75, 0x65, 0x73, 0x74, 0x12, 0x16, 0x0a, 0x06, 0x68,
-	0x65, 0x69, 0x67, 0x68, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28, 0x05, 0x52, 0x06, 0x68, 0x65, 0x69,
-	0x67, 0x68, 0x74, 0x12, 0x14, 0x0a, 0x05, 0x77, 0x69, 0x64, 0x74, 0x68, 0x18, 0x02, 0x20, 0x01,
-	0x28, 0x05, 0x52, 0x05, 0x77, 0x69, 0x64, 0x74, 0x68, 0x12, 0x12, 0x0a, 0x04, 0x6d, 0x6f, 0x64,
-	0x65, 0x18, 0x03, 0x20, 0x01, 0x28, 0x05, 0x52, 0x04, 0x6d, 0x6f, 0x64, 0x65, 0x12, 0x12, 0x0a,
-	0x04, 0x73, 0x74, 0x65, 0x70, 0x18, 0x04, 0x20, 0x01, 0x28, 0x05, 0x52, 0x04, 0x73, 0x74, 0x65,
-	0x70, 0x12, 0x12, 0x0a, 0x04, 0x73, 0x65, 0x65, 0x64, 0x18, 0x05, 0x20, 0x01, 0x28, 0x05, 0x52,
-	0x04, 0x73, 0x65, 0x65, 0x64, 0x12, 0x27, 0x0a, 0x0f, 0x70, 0x6f, 0x73, 0x69, 0x74, 0x69, 0x76,
-	0x65, 0x5f, 0x70, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x18, 0x06, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0e,
-	0x70, 0x6f, 0x73, 0x69, 0x74, 0x69, 0x76, 0x65, 0x50, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x12, 0x27,
-	0x0a, 0x0f, 0x6e, 0x65, 0x67, 0x61, 0x74, 0x69, 0x76, 0x65, 0x5f, 0x70, 0x72, 0x6f, 0x6d, 0x70,
-	0x74, 0x18, 0x07, 0x20, 0x01, 0x28, 0x09, 0x52, 0x0e, 0x6e, 0x65, 0x67, 0x61, 0x74, 0x69, 0x76,
-	0x65, 0x50, 0x72, 0x6f, 0x6d, 0x70, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x64, 0x73, 0x74, 0x18, 0x08,
-	0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x64, 0x73, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x73, 0x72, 0x63,
-	0x18, 0x09, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x73, 0x72, 0x63, 0x12, 0x2a, 0x0a, 0x10, 0x45,
-	0x6e, 0x61, 0x62, 0x6c, 0x65, 0x50, 0x61, 0x72, 0x61, 0x6d, 0x65, 0x74, 0x65, 0x72, 0x73, 0x18,
-	0x0a, 0x20, 0x01, 0x28, 0x09, 0x52, 0x10, 0x45, 0x6e, 0x61, 0x62, 0x6c, 0x65, 0x50, 0x61, 0x72,
-	0x61, 0x6d, 0x65, 0x74, 0x65, 0x72, 0x73, 0x12, 0x1a, 0x0a, 0x08, 0x43, 0x4c, 0x49, 0x50, 0x53,
-	0x6b, 0x69, 0x70, 0x18, 0x0b, 0x20, 0x01, 0x28, 0x05, 0x52, 0x08, 0x43, 0x4c, 0x49, 0x50, 0x53,
-	0x6b, 0x69, 0x70, 0x22, 0x48, 0x0a, 0x0a, 0x54, 0x54, 0x53, 0x52, 0x65, 0x71, 0x75, 0x65, 0x73,
-	0x74, 0x12, 0x12, 0x0a, 0x04, 0x74, 0x65, 0x78, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52,
-	0x04, 0x74, 0x65, 0x78, 0x74, 0x12, 0x14, 0x0a, 0x05, 0x6d, 0x6f, 0x64, 0x65, 0x6c, 0x18, 0x02,
-	0x20, 0x01, 0x28, 0x09, 0x52, 0x05, 0x6d, 0x6f, 0x64, 0x65, 0x6c, 0x12, 0x10, 0x0a, 0x03, 0x64,
-	0x73, 0x74, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x64, 0x73, 0x74, 0x22, 0x46, 0x0a,
-	0x14, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x61, 0x74, 0x69, 0x6f, 0x6e, 0x52, 0x65, 0x73,
-	0x70, 0x6f, 0x6e, 0x73, 0x65, 0x12, 0x16, 0x0a, 0x06, 0x6c, 0x65, 0x6e, 0x67, 0x74, 0x68, 0x18,
-	0x01, 0x20, 0x01, 0x28, 0x05, 0x52, 0x06, 0x6c, 0x65, 0x6e, 0x67, 0x74, 0x68, 0x12, 0x16, 0x0a,
-	0x06, 0x74, 0x6f, 0x6b, 0x65, 0x6e, 0x73, 0x18, 0x02, 0x20, 0x03, 0x28, 0x05, 0x52, 0x06, 0x74,
-	0x6f, 0x6b, 0x65, 0x6e, 0x73, 0x22, 0xac, 0x01, 0x0a, 0x0f, 0x4d, 0x65, 0x6d, 0x6f, 0x72, 0x79,
-	0x55, 0x73, 0x61, 0x67, 0x65, 0x44, 0x61, 0x74, 0x61, 0x12, 0x14, 0x0a, 0x05, 0x74, 0x6f, 0x74,
-	0x61, 0x6c, 0x18, 0x01, 0x20, 0x01, 0x28, 0x04, 0x52, 0x05, 0x74, 0x6f, 0x74, 0x61, 0x6c, 0x12,
-	0x45, 0x0a, 0x09, 0x62, 0x72, 0x65, 0x61, 0x6b, 0x64, 0x6f, 0x77, 0x6e, 0x18, 0x02, 0x20, 0x03,
-	0x28, 0x0b, 0x32, 0x27, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x4d, 0x65, 0x6d,
-	0x6f, 0x72, 0x79, 0x55, 0x73, 0x61, 0x67, 0x65, 0x44, 0x61, 0x74, 0x61, 0x2e, 0x42, 0x72, 0x65,
-	0x61, 0x6b, 0x64, 0x6f, 0x77, 0x6e, 0x45, 0x6e, 0x74, 0x72, 0x79, 0x52, 0x09, 0x62, 0x72, 0x65,
-	0x61, 0x6b, 0x64, 0x6f, 0x77, 0x6e, 0x1a, 0x3c, 0x0a, 0x0e, 0x42, 0x72, 0x65, 0x61, 0x6b, 0x64,
-	0x6f, 0x77, 0x6e, 0x45, 0x6e, 0x74, 0x72, 0x79, 0x12, 0x10, 0x0a, 0x03, 0x6b, 0x65, 0x79, 0x18,
-	0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x6b, 0x65, 0x79, 0x12, 0x14, 0x0a, 0x05, 0x76, 0x61,
-	0x6c, 0x75, 0x65, 0x18, 0x02, 0x20, 0x01, 0x28, 0x04, 0x52, 0x05, 0x76, 0x61, 0x6c, 0x75, 0x65,
-	0x3a, 0x02, 0x38, 0x01, 0x22, 0xbc, 0x01, 0x0a, 0x0e, 0x53, 0x74, 0x61, 0x74, 0x75, 0x73, 0x52,
-	0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73, 0x65, 0x12, 0x33, 0x0a, 0x05, 0x73, 0x74, 0x61, 0x74, 0x65,
-	0x18, 0x01, 0x20, 0x01, 0x28, 0x0e, 0x32, 0x1d, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64,
-	0x2e, 0x53, 0x74, 0x61, 0x74, 0x75, 0x73, 0x52, 0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73, 0x65, 0x2e,
-	0x53, 0x74, 0x61, 0x74, 0x65, 0x52, 0x05, 0x73, 0x74, 0x61, 0x74, 0x65, 0x12, 0x30, 0x0a, 0x06,
-	0x6d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x18, 0x02, 0x20, 0x01, 0x28, 0x0b, 0x32, 0x18, 0x2e, 0x62,
-	0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x4d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x55, 0x73, 0x61,
-	0x67, 0x65, 0x44, 0x61, 0x74, 0x61, 0x52, 0x06, 0x6d, 0x65, 0x6d, 0x6f, 0x72, 0x79, 0x22, 0x43,
-	0x0a, 0x05, 0x53, 0x74, 0x61, 0x74, 0x65, 0x12, 0x11, 0x0a, 0x0d, 0x55, 0x4e, 0x49, 0x4e, 0x49,
-	0x54, 0x49, 0x41, 0x4c, 0x49, 0x5a, 0x45, 0x44, 0x10, 0x00, 0x12, 0x08, 0x0a, 0x04, 0x42, 0x55,
-	0x53, 0x59, 0x10, 0x01, 0x12, 0x09, 0x0a, 0x05, 0x52, 0x45, 0x41, 0x44, 0x59, 0x10, 0x02, 0x12,
-	0x12, 0x0a, 0x05, 0x45, 0x52, 0x52, 0x4f, 0x52, 0x10, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
-	0xff, 0xff, 0x01, 0x32, 0xf4, 0x04, 0x0a, 0x07, 0x42, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x12,
-	0x32, 0x0a, 0x06, 0x48, 0x65, 0x61, 0x6c, 0x74, 0x68, 0x12, 0x16, 0x2e, 0x62, 0x61, 0x63, 0x6b,
-	0x65, 0x6e, 0x64, 0x2e, 0x48, 0x65, 0x61, 0x6c, 0x74, 0x68, 0x4d, 0x65, 0x73, 0x73, 0x61, 0x67,
-	0x65, 0x1a, 0x0e, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65, 0x70, 0x6c,
-	0x79, 0x22, 0x00, 0x12, 0x34, 0x0a, 0x07, 0x50, 0x72, 0x65, 0x64, 0x69, 0x63, 0x74, 0x12, 0x17,
-	0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x50, 0x72, 0x65, 0x64, 0x69, 0x63, 0x74,
-	0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x1a, 0x0e, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e,
-	0x64, 0x2e, 0x52, 0x65, 0x70, 0x6c, 0x79, 0x22, 0x00, 0x12, 0x35, 0x0a, 0x09, 0x4c, 0x6f, 0x61,
-	0x64, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x12, 0x15, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64,
-	0x2e, 0x4d, 0x6f, 0x64, 0x65, 0x6c, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x1a, 0x0f, 0x2e,
-	0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x22, 0x00,
-	0x12, 0x3c, 0x0a, 0x0d, 0x50, 0x72, 0x65, 0x64, 0x69, 0x63, 0x74, 0x53, 0x74, 0x72, 0x65, 0x61,
-	0x6d, 0x12, 0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x50, 0x72, 0x65, 0x64,
-	0x69, 0x63, 0x74, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x1a, 0x0e, 0x2e, 0x62, 0x61, 0x63,
-	0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65, 0x70, 0x6c, 0x79, 0x22, 0x00, 0x30, 0x01, 0x12, 0x40,
-	0x0a, 0x09, 0x45, 0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x12, 0x17, 0x2e, 0x62, 0x61,
-	0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x50, 0x72, 0x65, 0x64, 0x69, 0x63, 0x74, 0x4f, 0x70, 0x74,
-	0x69, 0x6f, 0x6e, 0x73, 0x1a, 0x18, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x45,
-	0x6d, 0x62, 0x65, 0x64, 0x64, 0x69, 0x6e, 0x67, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x22, 0x00,
-	0x12, 0x41, 0x0a, 0x0d, 0x47, 0x65, 0x6e, 0x65, 0x72, 0x61, 0x74, 0x65, 0x49, 0x6d, 0x61, 0x67,
-	0x65, 0x12, 0x1d, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x47, 0x65, 0x6e, 0x65,
-	0x72, 0x61, 0x74, 0x65, 0x49, 0x6d, 0x61, 0x67, 0x65, 0x52, 0x65, 0x71, 0x75, 0x65, 0x73, 0x74,
-	0x1a, 0x0f, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65, 0x73, 0x75, 0x6c,
-	0x74, 0x22, 0x00, 0x12, 0x4d, 0x0a, 0x12, 0x41, 0x75, 0x64, 0x69, 0x6f, 0x54, 0x72, 0x61, 0x6e,
-	0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x12, 0x1a, 0x2e, 0x62, 0x61, 0x63, 0x6b,
-	0x65, 0x6e, 0x64, 0x2e, 0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x52, 0x65,
-	0x71, 0x75, 0x65, 0x73, 0x74, 0x1a, 0x19, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e,
-	0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74,
-	0x22, 0x00, 0x12, 0x2d, 0x0a, 0x03, 0x54, 0x54, 0x53, 0x12, 0x13, 0x2e, 0x62, 0x61, 0x63, 0x6b,
-	0x65, 0x6e, 0x64, 0x2e, 0x54, 0x54, 0x53, 0x52, 0x65, 0x71, 0x75, 0x65, 0x73, 0x74, 0x1a, 0x0f,
-	0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x22,
-	0x00, 0x12, 0x4a, 0x0a, 0x0e, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x65, 0x53, 0x74, 0x72,
-	0x69, 0x6e, 0x67, 0x12, 0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x50, 0x72,
-	0x65, 0x64, 0x69, 0x63, 0x74, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x1a, 0x1d, 0x2e, 0x62,
-	0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x69, 0x7a, 0x61, 0x74,
-	0x69, 0x6f, 0x6e, 0x52, 0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73, 0x65, 0x22, 0x00, 0x12, 0x3b, 0x0a,
-	0x06, 0x53, 0x74, 0x61, 0x74, 0x75, 0x73, 0x12, 0x16, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e,
-	0x64, 0x2e, 0x48, 0x65, 0x61, 0x6c, 0x74, 0x68, 0x4d, 0x65, 0x73, 0x73, 0x61, 0x67, 0x65, 0x1a,
-	0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x53, 0x74, 0x61, 0x74, 0x75, 0x73,
-	0x52, 0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73, 0x65, 0x22, 0x00, 0x42, 0x5a, 0x0a, 0x19, 0x69, 0x6f,
-	0x2e, 0x73, 0x6b, 0x79, 0x6e, 0x65, 0x74, 0x2e, 0x6c, 0x6f, 0x63, 0x61, 0x6c, 0x61, 0x69, 0x2e,
-	0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x42, 0x0e, 0x4c, 0x6f, 0x63, 0x61, 0x6c, 0x41, 0x49,
-	0x42, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x50, 0x01, 0x5a, 0x2b, 0x67, 0x69, 0x74, 0x68, 0x75,
-	0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x67, 0x6f, 0x2d, 0x73, 0x6b, 0x79, 0x6e, 0x65, 0x74, 0x2f,
-	0x4c, 0x6f, 0x63, 0x61, 0x6c, 0x41, 0x49, 0x2f, 0x70, 0x6b, 0x67, 0x2f, 0x67, 0x72, 0x70, 0x63,
-	0x2f, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x06, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x33,
+	0x72, 0x69, 0x70, 0x74, 0x52, 0x65, 0x71, 0x75, 0x65, 0x73, 0x74, 0x1a, 0x19, 0x2e, 0x62, 0x61,
+	0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x54, 0x72, 0x61, 0x6e, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74,
+	0x52, 0x65, 0x73, 0x75, 0x6c, 0x74, 0x22, 0x00, 0x12, 0x2d, 0x0a, 0x03, 0x54, 0x54, 0x53, 0x12,
+	0x13, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x54, 0x54, 0x53, 0x52, 0x65, 0x71,
+	0x75, 0x65, 0x73, 0x74, 0x1a, 0x0f, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x52,
+	0x65, 0x73, 0x75, 0x6c, 0x74, 0x22, 0x00, 0x12, 0x4a, 0x0a, 0x0e, 0x54, 0x6f, 0x6b, 0x65, 0x6e,
+	0x69, 0x7a, 0x65, 0x53, 0x74, 0x72, 0x69, 0x6e, 0x67, 0x12, 0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b,
+	0x65, 0x6e, 0x64, 0x2e, 0x50, 0x72, 0x65, 0x64, 0x69, 0x63, 0x74, 0x4f, 0x70, 0x74, 0x69, 0x6f,
+	0x6e, 0x73, 0x1a, 0x1d, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x54, 0x6f, 0x6b,
+	0x65, 0x6e, 0x69, 0x7a, 0x61, 0x74, 0x69, 0x6f, 0x6e, 0x52, 0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73,
+	0x65, 0x22, 0x00, 0x12, 0x3b, 0x0a, 0x06, 0x53, 0x74, 0x61, 0x74, 0x75, 0x73, 0x12, 0x16, 0x2e,
+	0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e, 0x48, 0x65, 0x61, 0x6c, 0x74, 0x68, 0x4d, 0x65,
+	0x73, 0x73, 0x61, 0x67, 0x65, 0x1a, 0x17, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x2e,
+	0x53, 0x74, 0x61, 0x74, 0x75, 0x73, 0x52, 0x65, 0x73, 0x70, 0x6f, 0x6e, 0x73, 0x65, 0x22, 0x00,
+	0x42, 0x5a, 0x0a, 0x19, 0x69, 0x6f, 0x2e, 0x73, 0x6b, 0x79, 0x6e, 0x65, 0x74, 0x2e, 0x6c, 0x6f,
+	0x63, 0x61, 0x6c, 0x61, 0x69, 0x2e, 0x62, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x42, 0x0e, 0x4c,
+	0x6f, 0x63, 0x61, 0x6c, 0x41, 0x49, 0x42, 0x61, 0x63, 0x6b, 0x65, 0x6e, 0x64, 0x50, 0x01, 0x5a,
+	0x2b, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x67, 0x6f, 0x2d, 0x73,
+	0x6b, 0x79, 0x6e, 0x65, 0x74, 0x2f, 0x4c, 0x6f, 0x63, 0x61, 0x6c, 0x41, 0x49, 0x2f, 0x70, 0x6b,
+	0x67, 0x2f, 0x67, 0x72, 0x70, 0x63, 0x2f, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x06, 0x70, 0x72,
+	0x6f, 0x74, 0x6f, 0x33,
 }

 var (
--- a/pkg/grpc/proto/backend.proto
+++ b/pkg/grpc/proto/backend.proto
@@ -64,6 +64,7 @@ message PredictOptions {
  float NegativePromptScale = 39;
  string NegativePrompt = 40;
  int32 NDraft = 41;
+  repeated string Images = 42;
 }

 // The response message containing the result
@@ -116,6 +117,8 @@ message ModelOptions {
  // LLM (llama.cpp)
  string LoraBase = 35;
  string LoraAdapter = 36;
+  float LoraScale = 42;
+
  bool NoMulMatQ = 37;
  string DraftModel = 39;
  
@@ -123,6 +126,14 @@ message ModelOptions {

  // vllm
  string Quantization = 40;
+
+  string MMProj = 41;
+
+  string RopeScaling = 43;
+  float YarnExtFactor = 44;
+  float YarnAttnFactor = 45;
+  float YarnBetaFast = 46;
+  float YarnBetaSlow = 47;
 }

 message Result {
--- a/pkg/grpc/proto/backend_grpc.pb.go
+++ b/pkg/grpc/proto/backend_grpc.pb.go
@@ -1,7 +1,7 @@
 // Code generated by protoc-gen-go-grpc. DO NOT EDIT.
 // versions:
 // - protoc-gen-go-grpc v1.2.0
-// - protoc             v3.15.8
+// - protoc             v4.23.4
 // source: pkg/grpc/proto/backend.proto

 package proto
--- a/pkg/model/initializers.go
+++ b/pkg/model/initializers.go
@@ -63,7 +63,7 @@ var AutoLoadBackends []string = []string{
 // It also loads the model
 func (ml *ModelLoader) grpcModel(backend string, o *Options) func(string, string) (*grpc.Client, error) {
 	return func(modelName, modelFile string) (*grpc.Client, error) {
-		log.Debug().Msgf("Loading GRPC Model %s: %+v", backend, *o)
+		log.Debug().Msgf("Loading Model %s with gRPC (file: %s) (backend: %s): %+v", modelName, modelFile, backend, *o)

 		var client *grpc.Client
Author	SHA1	Message	Date
Ettore Di Giacinto	a8e91345e2	fixup: create piper libdir also when not built Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-11-12 22:17:11 +01:00
Ettore Di Giacinto	ea4ade6b60	Merge branch 'master' into enable_gpu	2023-11-11 19:20:36 +01:00
Ettore Di Giacinto	803a0ac02a	feat(llama.cpp): support lora with scale and yarn (#1277 ) * feat(llama.cpp): support lora with scale Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(llama.cpp): support yarn Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-11-11 18:40:48 +01:00
Gianluca Boiano	bde87d00b9	deps(go-piper): update to 2023.11.6-3 (#1257 ) Signed-off-by: Gianluca Boiano <morf3089@gmail.com>	2023-11-11 18:40:26 +01:00
Ettore Di Giacinto	0eae727366	🔥 add LaVA support and GPT vision API, Multiple requests for llama.cpp, return JSON types (#1254 ) * wip * wip * Make it functional Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * wip * Small fixups * do not inject space on role encoding, encode img at beginning of messages Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Add examples/config defaults * Add include dir of current source dir * cleanup * fixes Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fixups * Revert "fixups" This reverts commit `f1a4731cca`. * fixes Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-11-11 13:14:59 +01:00
LocalAI [bot]	3b4c5d54d8	⬆️ Update ggerganov/llama.cpp (#1265 ) Signed-off-by: GitHub <noreply@github.com> Co-authored-by: mudler <mudler@users.noreply.github.com>	2023-11-10 08:50:42 +01:00
LocalAI [bot]	4e16bc2f13	⬆️ Update ggerganov/llama.cpp (#1256 ) Signed-off-by: GitHub <noreply@github.com> Co-authored-by: mudler <mudler@users.noreply.github.com>	2023-11-08 08:21:12 +01:00
LocalAI [bot]	562ac62f59	⬆️ Update ggerganov/llama.cpp (#1242 ) Signed-off-by: GitHub <noreply@github.com> Co-authored-by: mudler <mudler@users.noreply.github.com>	2023-11-07 08:37:55 +01:00
Ettore Di Giacinto	796d0c99aa	tests: re-enable GPU tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-11-06 19:10:35 +01:00
Diego	e7fa2e06f8	Fixes the bug 1196 (#1232 ) * Current state of the branch. * Now gRPC is build only when the BUILD_GRPC_FOR_BACKEND_LLAMA variable is defined. * Now the local compilation of gRPC is executed on BUILD_GRPC_FOR_BACKEND_LLAMA. * Revised the Makefile. * Removed replace directives in go.mod. --------- Signed-off-by: Diego <38375572+diego-minguzzi@users.noreply.github.com> Co-authored-by: lunamidori5 <118759930+lunamidori5@users.noreply.github.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>	2023-11-06 19:07:46 +01:00
Ettore Di Giacinto	8123f009d0	dockerfile: fixup duplicate This should have been "exllama" Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-11-05 14:09:31 +01:00
Ettore Di Giacinto	622aaa9f7d	dockerfile: avoid pushing a big layer Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-11-05 10:31:33 +01:00
Ettore Di Giacinto	7b1ee203ce	tests: re-add flake-attempts Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-11-05 09:01:03 +01:00
Ettore Di Giacinto	f347e51927	feat(conda): conda environments (#1144 ) * feat(autogptq): add a separate conda environment for autogptq (#1137) Description This PR related to #1117 Notes for Reviewers Here we lock down the version of the dependencies. Make sure it can be used all the time without failed if the version of dependencies were upgraded. I change the order of importing packages according to the pylint, and no change the logic of code. It should be ok. I will do more investigate on writing some test cases for every backend. I can run the service in my environment, but there is not exist a way to test it. So, I am not confident on it. Add a README.md in the `grpc` root. This is the common commands for creating `conda` environment. And it can be used to the reference file for creating extral gRPC backend document. Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * [Extra backend] Add seperate environment for ttsbark (#1141) Description This PR relates to #1117 Notes for Reviewers Same to the latest PR: * The code is also changed, but only the order of the import package parts. And some code comments are also added. * Add a configuration of the `conda` environment * Add a simple test case for testing if the service can be startup in current `conda` environment. It is succeed in VSCode, but the it is not out of box on terminal. So, it is hard to say the test case really useful. [Signed commits](../CONTRIBUTING.md#signing-off-on-commits-developer-certificate-of-origin) - [x] Yes, I signed my commits. <!-- Thank you for contributing to LocalAI! Contributing Conventions ------------------------- The draft above helps to give a quick overview of your PR. Remember to remove this comment and to at least: 1. Include descriptive PR titles with [<component-name>] prepended. We use [conventional commits](https://www.conventionalcommits.org/en/v1.0.0/). 2. Build and test your changes before submitting a PR (`make build`). 3. Sign your commits 4. Tag maintainer: for a quicker response, tag the relevant maintainer (see below). 5. X/Twitter handle: we announce bigger features on X/Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! By following the community's contribution conventions upfront, the review process will be accelerated and your PR merged more quickly. If no one reviews your PR within a few days, please @-mention @mudler. --> Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(conda): add make target and entrypoints for the dockerfile Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(conda): Add seperate conda env for diffusers (#1145) Description This PR relates to #1117 Notes for Reviewers * Add `conda` env `diffusers.yml` * Add Makefile to create it automatically * Add `run.sh` to support running as a extra backend * Also adding it to the main Dockerfile * Add make command in the root Makefile * Testing the server, it can start up under the env Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(conda):Add seperate env for vllm (#1148) Description This PR is related to #1117 Notes for Reviewers * The gRPC server can be started as normal * The test case can be triggered in VSCode * Same to other this kind of PRs, add `vllm.yml` Makefile and add `run.sh` to the main Dockerfile, and command to the main Makefile [Signed commits](../CONTRIBUTING.md#signing-off-on-commits-developer-certificate-of-origin) - [x] Yes, I signed my commits. <!-- Thank you for contributing to LocalAI! Contributing Conventions ------------------------- The draft above helps to give a quick overview of your PR. Remember to remove this comment and to at least: 1. Include descriptive PR titles with [<component-name>] prepended. We use [conventional commits](https://www.conventionalcommits.org/en/v1.0.0/). 2. Build and test your changes before submitting a PR (`make build`). 3. Sign your commits 4. Tag maintainer: for a quicker response, tag the relevant maintainer (see below). 5. X/Twitter handle: we announce bigger features on X/Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! By following the community's contribution conventions upfront, the review process will be accelerated and your PR merged more quickly. If no one reviews your PR within a few days, please @-mention @mudler. --> Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(conda):Add seperate env for huggingface (#1146) Description This PR is related to #1117 Notes for Reviewers * Add conda env `huggingface.yml` * Change the import order, and also remove the no-used packages * Add `run.sh` and `make command` to the main Dockerfile and Makefile * Add test cases for it. It can be triggered and succeed under VSCode Python extension but it is hang by using `python -m unites test_huggingface.py` in the terminal ``` Running tests (unittest): /workspaces/LocalAI/extra/grpc/huggingface Running tests: /workspaces/LocalAI/extra/grpc/huggingface/test_huggingface.py::TestBackendServicer::test_embedding /workspaces/LocalAI/extra/grpc/huggingface/test_huggingface.py::TestBackendServicer::test_load_model /workspaces/LocalAI/extra/grpc/huggingface/test_huggingface.py::TestBackendServicer::test_server_startup ./test_huggingface.py::TestBackendServicer::test_embedding Passed ./test_huggingface.py::TestBackendServicer::test_load_model Passed ./test_huggingface.py::TestBackendServicer::test_server_startup Passed Total number of tests expected to run: 3 Total number of tests run: 3 Total number of tests passed: 3 Total number of tests failed: 0 Total number of tests failed with errors: 0 Total number of tests skipped: 0 Finished running tests! ``` [Signed commits](../CONTRIBUTING.md#signing-off-on-commits-developer-certificate-of-origin) - [x] Yes, I signed my commits. <!-- Thank you for contributing to LocalAI! Contributing Conventions ------------------------- The draft above helps to give a quick overview of your PR. Remember to remove this comment and to at least: 1. Include descriptive PR titles with [<component-name>] prepended. We use [conventional commits](https://www.conventionalcommits.org/en/v1.0.0/). 2. Build and test your changes before submitting a PR (`make build`). 3. Sign your commits 4. Tag maintainer: for a quicker response, tag the relevant maintainer (see below). 5. X/Twitter handle: we announce bigger features on X/Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! By following the community's contribution conventions upfront, the review process will be accelerated and your PR merged more quickly. If no one reviews your PR within a few days, please @-mention @mudler. --> Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(conda): Add the seperate conda env for VALL-E X (#1147) Description This PR is related to #1117 Notes for Reviewers * The gRPC server cannot start up ``` (ttsvalle) @Aisuko ➜ /workspaces/LocalAI (feat/vall-e-x) $ /opt/conda/envs/ttsvalle/bin/python /workspaces/LocalAI/extra/grpc/vall-e-x/ttsvalle.py Traceback (most recent call last): File "/workspaces/LocalAI/extra/grpc/vall-e-x/ttsvalle.py", line 14, in <module> from utils.generation import SAMPLE_RATE, generate_audio, preload_models ModuleNotFoundError: No module named 'utils' ``` The installation steps follow https://github.com/Plachtaa/VALL-E-X#-installation below: * Under the `ttsvalle` conda env ``` git clone https://github.com/Plachtaa/VALL-E-X.git cd VALL-E-X pip install -r requirements.txt ``` [Signed commits](../CONTRIBUTING.md#signing-off-on-commits-developer-certificate-of-origin) - [x] Yes, I signed my commits. <!-- Thank you for contributing to LocalAI! Contributing Conventions ------------------------- The draft above helps to give a quick overview of your PR. Remember to remove this comment and to at least: 1. Include descriptive PR titles with [<component-name>] prepended. We use [conventional commits](https://www.conventionalcommits.org/en/v1.0.0/). 2. Build and test your changes before submitting a PR (`make build`). 3. Sign your commits 4. Tag maintainer: for a quicker response, tag the relevant maintainer (see below). 5. X/Twitter handle: we announce bigger features on X/Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! By following the community's contribution conventions upfront, the review process will be accelerated and your PR merged more quickly. If no one reviews your PR within a few days, please @-mention @mudler. --> Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix: set image type Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(conda):Add seperate conda env for exllama (#1149) Add seperate env for exllama Signed-off-by: Aisuko <urakiny@gmail.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Setup conda Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Set image_type arg Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci: prepare only conda env in tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Dockerfile: comment manual pip calls Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * conda: add conda to PATH Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fixes * add shebang * Fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * file perms Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * debug * Install new conda in the worker * Disable GPU tests for now until the worker is back * Rename workflows * debug * Fixup conda install * fixup(wrapper): pass args Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Signed-off-by: Aisuko <urakiny@gmail.com> Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com> Co-authored-by: Aisuko <urakiny@gmail.com>	2023-11-04 15:30:32 +01:00
LocalAI [bot]	9b17af18b3	⬆️ Update ggerganov/llama.cpp (#1236 ) Signed-off-by: GitHub <noreply@github.com> Co-authored-by: mudler <mudler@users.noreply.github.com>	2023-11-03 19:23:53 +01:00
Samuel Walker	23c7fbfe6b	chianlit example (#1238 )	2023-11-02 22:56:46 +01:00
Samuel Walker	035fea676a	llama index example (#1237 )	2023-11-02 13:35:06 -07:00
Vitor Oliveira	6e1a234d15	feat(certificates): add support for custom CA certificates (#880 ) This change facilitates users working behind corporate firewalls or proxies. By allowing the integration of custom CA certificates, users can handle SSL connections that are intercepted by company infrastructure.	2023-11-01 20:10:14 +01:00
LocalAI [bot]	5b596ea605	⬆️ Update ggerganov/llama.cpp (#1231 ) Signed-off-by: GitHub <noreply@github.com> Co-authored-by: mudler <mudler@users.noreply.github.com>	2023-11-01 12:44:34 +00:00
Dave	6bd56460de	Update .gitignore for backend/llama.cpp (#1235 ) Signed-off-by: Dave <dave@gray101.com>	2023-11-01 09:52:02 +01:00
LocalAI [bot]	6ef7ea2635	⬆️ Update ggerganov/llama.cpp (#1207 ) Signed-off-by: GitHub <noreply@github.com> Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com> Co-authored-by: mudler <mudler@users.noreply.github.com>	2023-10-30 08:00:36 +00:00
Ettore Di Giacinto	f8c00fbaf1	ci: enlarge download timeout window Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-10-29 22:09:35 +01:00
Ettore Di Giacinto	d9a42cc4c5	ci: run only cublas on selfhosted (#1224 ) * ci: run only cublas on selfhosted Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * debug Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * update git Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * change testing embeddings model link Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2023-10-29 22:04:43 +01:00
Ettore Di Giacinto	fc0bc32814	ci: use self-hosted to build container images (#1206 ) ci: use self-hosted Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>	2023-10-26 21:13:40 +02:00