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 supported59 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix][Quantization] Honor activation scheme in MXFP8 linear selection
- [Bugfix][Core] Fail requests queued during level-2 sleep when a wake resumes before the weights are restored
- [Bug]: Large `n` (parallel sampling) in /v1/chat/completions causes severe cross-request latency
- [Feature][LoRA] Coordinate MoE adapter admission and residency
- Share DeepSeek V4.1 CPU KV across TP ranks
- [Bug][Quantization] MXFP8 linear kernel selection ignores the requested activation scheme — weight-only (W8A16) checkpoints silently run W8A8
- [ROCm] Move AITER MXFP4 linear op registration into _aiter_ops
- [Quantization] OSCAR KV cache presets for the TurboQuant backend
- [Core] Priority scheduling: hold lower-priority prefill for first tokens
- [Core] DRYup KV cache block allocation loop
- Docs
- Python not yet supported