biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Docker] Add fastsafetensors to NVIDIA Dockerfile (#38950)

Browse Source
This commit is contained in:
Zhewen Li
2026-04-08 22:21:37 -07:00
committed by GitHub
parent e80e633927
commit 9e78555743
5 changed files with 14 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docker/Dockerfile
View File

@@ -538,7 +538,9 @@ RUN CUDA_VERSION_DASH=$(echo $CUDA_VERSION | cut -d. -f1,2 | tr '.' '-') && \
cuda-nvrtc-${CUDA_VERSION_DASH} \
cuda-cuobjdump-${CUDA_VERSION_DASH} \
libcurand-dev-${CUDA_VERSION_DASH} \
libcublas-${CUDA_VERSION_DASH} && \
libcublas-${CUDA_VERSION_DASH} \
# Required by fastsafetensors (fixes #20384)
libnuma-dev && \
# Fixes nccl_allocator requiring nccl.h at runtime
# https://github.com/vllm-project/vllm/blob/1336a1ea244fa8bfd7e72751cabbdb5b68a0c11a/vllm/distributed/device_communicators/pynccl_allocator.py#L22
# NCCL packages don't use the cuda-MAJOR-MINOR naming convention,