Mirrors
/
hf_text-generation-inference
mirror of https://github.com/huggingface/text-generation-inference.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

hotfix: update nccl

This commit is contained in:
OlivierDehaene 2024-07-23 23:31:28 +02:00
parent e7e3aa6cac
commit a895029424
No known key found for this signature in database
GPG Key ID: BB104D67809DA93C
2 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
Dockerfile
View File

@ -252,7 +252,10 @@ COPY server/Makefile server/Makefile
RUN cd server && \ RUN cd server && \
make gen-server && \ make gen-server && \
pip install -r requirements_cuda.txt && \ pip install -r requirements_cuda.txt && \
pip install ".[bnb, accelerate, quantize, peft, outlines]" --no-cache-dir pip install ".[bnb, accelerate, quantize, peft, outlines]" --no-cache-dir && \
pip install nvidia-nccl-cu12==2.22.3
ENV LD_PRELOAD=/opt/conda/lib/python3.10/site-packages/nvidia/nccl/lib/libnccl.so.2
# Deps before the binaries # Deps before the binaries
# The binaries change on every build given we burn the SHA into them # The binaries change on every build given we burn the SHA into them

1
server/Makefile
View File

@ -30,6 +30,7 @@ install: install-cuda
install-cuda: install-server install-flash-attention-v2-cuda install-vllm-cuda install-flash-attention install-fbgemm install-cuda: install-server install-flash-attention-v2-cuda install-vllm-cuda install-flash-attention install-fbgemm
pip install -e ".[bnb]" pip install -e ".[bnb]"
pip install nvidia-nccl-cu12==2.22.3
install-rocm: install-server install-flash-attention-v2-rocm install-vllm-rocm install-rocm: install-server install-flash-attention-v2-rocm install-vllm-rocm