dearwolf/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2025-05-28 14:35:00 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 12

feat: Add exllama (#881)

Browse source 
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2023-08-11 00:49:40 +02:00 • committed by GitHub
parent 1d1cae8e4d
commit ff3ab5fcca
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
4 changed files with 491 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

6
Dockerfile
View file

@ -11,7 +11,7 @@ ARG TARGETARCH
ARG TARGETVARIANT
ENV BUILD_TYPE=${BUILD_TYPE}
ENV EXTERNAL_GRPC_BACKENDS="huggingface-embeddings:/build/extra/grpc/huggingface/huggingface.py,autogptq:/build/extra/grpc/autogptq/autogptq.py,bark:/build/extra/grpc/bark/ttsbark.py,diffusers:/build/extra/grpc/diffusers/backend_diffusers.py"
ENV EXTERNAL_GRPC_BACKENDS="huggingface-embeddings:/build/extra/grpc/huggingface/huggingface.py,autogptq:/build/extra/grpc/autogptq/autogptq.py,bark:/build/extra/grpc/bark/ttsbark.py,diffusers:/build/extra/grpc/diffusers/backend_diffusers.py,exllama:/build/extra/grpc/exllama/exllama.py"
ARG GO_TAGS="stablediffusion tts"
RUN apt-get update && \
@ -29,7 +29,7 @@ RUN if [ "${BUILD_TYPE}" = "cublas" ]; then \
dpkg -i cuda-keyring_1.0-1_all.deb && \
rm -f cuda-keyring_1.0-1_all.deb && \
apt-get update && \
apt-get install -y cuda-nvcc-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} libcublas-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
apt-get install -y cuda-nvcc-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} libcublas-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} libcusparse-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} libcusolver-dev-${CUDA_MAJOR_VERSION}-${CUDA_MINOR_VERSION} \
; fi
ENV PATH /usr/local/cuda/bin:${PATH}
@ -42,7 +42,7 @@ RUN if [ "${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 && pip install auto-gptq;\
pip install torch && 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