A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[CI/Build] Update flashinfer to 0.2.9 (#22233)
Signed-off-by: mgoin <mgoin64@gmail.com>
Michael Goin committed
Aug 5, 2025 at 12:39 UTC
a7cb6101ca7bd3d3ee94a5fe37caab8ebca32d80
2 files changed
+2
-2
docker/Dockerfile
+1
-1
@@ -392,7 +392,7 @@ RUN --mount=type=bind,from=build,src=/workspace/dist,target=/vllm-workspace/dist
392
ARG FLASHINFER_GIT_REPO="https://github.com/flashinfer-ai/flashinfer.git"
393
# Keep this in sync with https://github.com/vllm-project/vllm/blob/main/requirements/cuda.txt
394
# We use `--force-reinstall --no-deps` to avoid issues with the existing FlashInfer wheel.
395
-ARG FLASHINFER_GIT_REF="v0.2.9rc2"
395
+ARG FLASHINFER_GIT_REF="v0.2.9"
396
RUN --mount=type=cache,target=/root/.cache/uv bash - <<'BASH'
397
. /etc/environment
398
git clone --depth 1 --recursive --shallow-submodules \
setup.py
+1
-1
@@ -665,7 +665,7 @@ setup(
665
"mistral_common[audio]"], # Required for audio processing
666
"video": [], # Kept for backwards compatibility
667
# FlashInfer should be updated together with the Dockerfile
668
- "flashinfer": ["flashinfer-python==0.2.9rc2"],
668
+ "flashinfer": ["flashinfer-python==0.2.9"],
669
},
670
cmdclass=cmdclass,
671
package_data=package_data,