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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Perf][Mamba] Capture the Mamba2 mixer in piecewise CUDA graphs
- [Bugfix][NIXL] Transfer Mamba2 conv state across TP sizes when B/C are replicated
- [Spec Decode][MRV2] Support heterogeneous-vocab draft models (TLI)
- [Frontend] Support cross-machine snapshot (step 1: portable attachments)
- [Bugfix][Core] Redo prompt logprobs for requests resumed after reset_prefix_cache
- [Docs][KV Connector] Document NIXL KV connector metrics aggregation semantics
- [Bugfix] Stop engines before API servers so --shutdown-timeout drains with --api-server-count > 1
- [XPU][Platform] Handle nodes with both CUDA and XPU drivers active
- [RFC]: Early KV transfer for direct Mooncake P/D
- [ROCm][Perf][Qwen3.8] Add dense mono decode for Qwen3.8-27B on MI355X.
- Docs
- Python not yet supported