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
- [Attention] Add P256 MiniMax M3 indexer CP layout
- [Fast Start] Support compressed-tensors CUTLASS FP8 weight-cache restarts
- [Bugfix][MoE] Fall back to torch routing for unsupported sqrtsoftplus expert counts
- [Feature][KV Connector] Pass typed cache axes at registration
- [Bugfix][MoE] Avoid double router weighting in TRTLLM BF16 experts
- [KV Connector] Add pause mode "migrate" for KV-preserving request handoff
- [Perf][Spec Decode] Skip redundant DSpark context projections for long prefills
- [Bugfix][ROCm] Take DSA candidate scores and sampler fp32 logits from persistent buffers
- [Bugfix][MRV2] Capture CUDA graphs with execute_model's input_ids
- [Perf][dsv4-flash-vision] Accelerate long-context C128 prefill on Hopper
- Docs
- Python not yet supported