huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-04-19 22:02:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore: update torch (#1730)

Browse Source 
Co-authored-by: Nicolas Patry <patry.nicolas@protonmail.com>
This commit is contained in:
OlivierDehaene 2024-04-30 14:04:28 +02:00 committed by GitHub
parent c99ecd77ec
commit b4ef038837
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
Dockerfile
View File

@ -39,7 +39,7 @@ RUN cargo build --release
# Adapted from: https://github.com/pytorch/pytorch/blob/master/Dockerfile # Adapted from: https://github.com/pytorch/pytorch/blob/master/Dockerfile
FROM nvidia/cuda:12.1.0-devel-ubuntu22.04 as pytorch-install FROM nvidia/cuda:12.1.0-devel-ubuntu22.04 as pytorch-install
ARG PYTORCH_VERSION=2.1.1 ARG PYTORCH_VERSION=2.3.0
ARG PYTHON_VERSION=3.10 ARG PYTHON_VERSION=3.10
# Keep in sync with `server/pyproject.toml # Keep in sync with `server/pyproject.toml
ARG CUDA_VERSION=12.1 ARG CUDA_VERSION=12.1
@ -149,6 +149,8 @@ FROM kernel-builder as vllm-builder
WORKDIR /usr/src WORKDIR /usr/src
ENV TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 8.9 9.0+PTX"
COPY server/Makefile-vllm Makefile COPY server/Makefile-vllm Makefile
# Build specific version of vllm # Build specific version of vllm
@ -210,7 +212,7 @@ COPY --from=vllm-builder /usr/src/vllm/build/lib.linux-x86_64-cpython-310 /opt/c
COPY --from=mamba-builder /usr/src/mamba/build/lib.linux-x86_64-cpython-310/ /opt/conda/lib/python3.10/site-packages COPY --from=mamba-builder /usr/src/mamba/build/lib.linux-x86_64-cpython-310/ /opt/conda/lib/python3.10/site-packages
COPY --from=mamba-builder /usr/src/causal-conv1d/build/lib.linux-x86_64-cpython-310/ /opt/conda/lib/python3.10/site-packages COPY --from=mamba-builder /usr/src/causal-conv1d/build/lib.linux-x86_64-cpython-310/ /opt/conda/lib/python3.10/site-packages
# Install vllm/flash-attention dependencies # Install flash-attention dependencies
RUN pip install einops --no-cache-dir RUN pip install einops --no-cache-dir
# Install server # Install server
@ -246,6 +248,7 @@ ENTRYPOINT ["./entrypoint.sh"]
FROM base FROM base
COPY ./tgi-entrypoint.sh /tgi-entrypoint.sh COPY ./tgi-entrypoint.sh /tgi-entrypoint.sh
RUN chmod +x /tgi-entrypoint.sh
ENTRYPOINT ["/tgi-entrypoint.sh"] ENTRYPOINT ["/tgi-entrypoint.sh"]
CMD ["--json-output"] CMD ["--json-output"]

2
server/Makefile-vllm
View File

@ -4,7 +4,7 @@ vllm-cuda:
git clone https://github.com/OlivierDehaene/vllm.git vllm git clone https://github.com/OlivierDehaene/vllm.git vllm
build-vllm-cuda: vllm-cuda build-vllm-cuda: vllm-cuda
cd vllm && git fetch && git checkout 4bec8cee87f6bb8cebaec297029713cd2082e0b2 cd vllm && git fetch && git checkout 3d4693536dcb69f036c26b016a35839b99ebed59
cd vllm && python setup.py build cd vllm && python setup.py build
install-vllm-cuda: build-vllm-cuda install-vllm-cuda: build-vllm-cuda