vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Perf] Fuse small Gemma3n AltUp residual updates with baddbmm
- [Perf] Enable efficient SDPA for Granite Speech block attention
- Fix/dsv4 sparse mla portability
- [Bug]: DFlash2 (dflash method) deterministic cumulative OOB — engine dies with CUDA IMA / Xid 31 after ~11k decode steps under sampling load (sm_80, v0.27.1)
- [Observability] Add bounded in-flight queue diagnostics
- [Bug]: Speculative decoding draft and target can select attention backends with disjoint KV cache layouts
- [Bugfix][DeepEP] Synchronize EP ranks before CUDA graph capture
- fix: handle multiple KV groups in _update_requests_with_invalid_blocks (HMA/hybrid models)
- [Kernel] Use FlashInfer's gemma RMSNorm kernels on CUDA
- [Bugfix] Fix compressed tensors MXFP4 MoE backend
- Docs
- Python not yet supported