vllm/requirements at c7b1080a64cd2d9aee2361d69ec00f9aff14027c - vllm

Files

T

Tyler Michael SmithandClaude Opus 4.6 7ed39716c2 [Build] Downgrade default CUDA version to 12.8 for PyTorch cu128 compatibility

PyTorch official PyPI wheels are built on CUDA 12.8, so pip install vllm
with a cu129 wheel pins nvidia-cublas-cu12 to 12.8 and causes a runtime
conflict (pytorch/pytorch#174949). Downgrade the default CUDA version so
the CI image, test image, and PyPI wheel all use cu128 consistently.

- Dockerfile: CUDA_VERSION 12.9.1 → 12.8.1
- requirements/test.txt: recompile lockfile with --torch-backend cu128
- .pre-commit-config.yaml: match pip-compile hook to cu128
- docker/versions.json: regenerated

Release builds (cu129, cu130) are unaffected since they pass explicit
--build-arg CUDA_VERSION overrides.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>

2026-02-18 17:59:30 -05:00

build.txt

Patch protobuf for CVE-2026-0994 (#34253 )

2026-02-11 02:25:04 -08:00

common.txt

[Voxtral Realtime] Refactor & Improve buffering logic (#34428 )

2026-02-12 09:46:43 -08:00

cpu-build.txt

[CPU][Feat] Update PyTorch to v2.10 for CPU Backend (#32869 )

2026-01-23 21:13:06 +08:00

cpu.txt

[CPU][Feat] Update PyTorch to v2.10 for CPU Backend (#32869 )