software-engineering/spark-vllm-docker
2
0
Fork 0
Code Issues Actions Packages Projects Releases Wiki Activity

Merge branch 'pr-152'

Browse Source
This commit is contained in:
eugr
2026-04-01 08:59:01 -07:00
parent ae25d64ac0 7b47235463
commit 3335540972
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
Dockerfile.mxfp4
View File

@@ -101,7 +101,7 @@ ARG FLASHINFER_SHA=f349e52496a72a00d8c4ac02c7a1e38523ff7194
ARG CUTLASS_SHA=fede53000a962b46e05bafe0c86311778caeb380
RUN --mount=type=cache,id=uv-cache,target=/root/.cache/uv \
uv pip install nvidia-nvshmem-cu13 "apache-tvm-ffi<0.2"
uv pip install "nvidia-nvshmem-cu13<3.6" "apache-tvm-ffi<0.2"
# Clone FlashInfer (cached for faster rebuilds)
RUN --mount=type=cache,id=git-flashinfer,target=/git-cache/flashinfer \
@@ -272,7 +272,7 @@ ENV PATH=$VLLM_BASE_DIR:$PATH
# Final extra deps
RUN --mount=type=cache,id=uv-cache,target=/root/.cache/uv \
uv pip install ray[default] fastsafetensors nvidia-nvshmem-cu13
uv pip install ray[default] fastsafetensors "nvidia-nvshmem-cu13<3.6"
# Build metadata (generated by build-and-copy.sh)
COPY build-metadata.yaml /workspace/build-metadata.yaml