feat(vllm): Upgrade to vLLM v0.25.0rc1#181
Closed
abatilo wants to merge 1 commit into
Closed
Conversation
Pin vllm-commit to b2dec4ac5a7942fdac7e687175bebba2f1d43575, the commit the v0.25.0rc1 tag points at. torch 2.11.0/vision 0.26.0/audio 2.11.0 requirements are unchanged upstream, so both CUDA base images stay the same, and FlashInfer v0.6.14 with the GLM routing patch remains ahead of upstream's 0.6.13 pin. Comment-only Dockerfile updates record that the quack-kernels pin was re-verified for 0.25 and that the llguidance pin now mirrors upstream's own requirement.
|
@abatilo Build complete, success: https://github.com/coreweave/ml-containers/actions/runs/28944020569 |
|
@abatilo Build complete, success: https://github.com/coreweave/ml-containers/actions/runs/28944020569 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
vllm-committob2dec4ac5a7942fdac7e687175bebba2f1d43575, the commit the v0.25.0rc1 tag points at, in both CUDA matrix entries (13.2.1 and 12.9.1).Compatibility verified at the new commit
CUDA_SUPPORTED_ARCHSlist still lacks 10.3, so the CMakeLists sed patch (and its grep guard) still applies.build_rust.shartifacts stillvllm/vllm-rs+vllm/_rust_*.so,tools/install_protoc.shpresent.Risk
v0.25 adds new runtime deps installed from PyPI in the final stage (tilelang, humming-kernels, PyNvVideoCodec, tokenspeed-mla, torchcodec, numba 0.65, fastsafetensors); if any lacks an arm64/cu12 wheel the build will fail at that step — watching CI for this.