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
- [Bugfix] Disable FlashInfer on sm_120 when the CUDA toolkit is older than 12.9
- [MoE] enable all2all fi_one_sided by default only for GB200/GB300
- [Bugfix][Distributed] Fail clearly when a confirmed local shm reader can't attach
- [Bug]: VLLM_DEEPEP_V2_ALLOW_HYBRID_MODE declares a default of True but resolves to False
- [Bugfix] Reject FP8 Triton MoE below SM89 instead of an opaque compile crash
- [Doc] Context extension: overrides must target text_config for composite models
- [Kernel] Refactor Int8ScaledMMLinearLayerConfig to use QuantKey (#32268)
- [Bugfix][Responses] Backfill required fields that arrive as explicit null
- [Docs] Document NIXL KV connector metrics aggregation semantics (#41230)
- [Bug]: Direct DCP A2A crashes on GLM sparse-MLA strided output
- Docs
- Python not yet supported