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
- [WIP] Handle ragged speculative decode in CPU ShortConv
- [WIP] Avoid pixel allocation when counting Pixtral image patches
- [WIP] Release unused LoRA resolver locks
- [WIP] Bound retained batch forward timing samples
- [WIP] Retain async worker capacity until cancelled jobs finish
- [WIP] Bound GLM video frame sampling work
- fix(renderers): preserve special tokens when truncation_side matches tokenizer default
- [Bug]: NixlConnector silently corrupts Mamba2 conv state under heterogeneous TP when n_groups is not divisible by TP (e.g. granite-4.0-h, n_groups=1)
- [Perf][Mamba] Capture the Mamba2 mixer in piecewise CUDA graphs
- [Bugfix][NIXL] Transfer Mamba2 conv state across TP sizes when B/C are replicated
- Docs
- Python not yet supported