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 supported43 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- fix(logging): use cyan for DEBUG level so logs stay legible on light terminals
- support pp+mtp & pp+eagle3
- Consolidate ZMQ context manager
- Skip stop string matching inside <think> blocks
- [MM][CG] Add user-configurable global/local token budgets for dual-path ViT CUDA graph
- [Bugfix] Fix hetero TP handshake assertion for GQA-replicated KV heads
- [Doc] Clarify prompt embeds token IDs in offline example
- [Feat] Support FlashQLA backend for qwen3.5
- [XPU][MiniMax-M3] Add XPU support for MSA sparse-attention path + tune gemma_rmsnorm for BMG
- [Performance]: TTFT of graph mode is about 25% worse than that of eager mode.
- Docs
- Python not yet supported