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
- [Bugfix][collect-env] Distinguish CUDA Toolkit and Runtime versions
- [Benchmark] Record model_id in bench startup --output-json
- [Bugfix][Model] Avoid rank-local compilation of GLM indexer key norm
- [Performance]: DeepSeek-V3.2 under DCP: FP8 sparse attention processes other ranks' masked top-k slots on Hopper
- [Benchmark] Record output token counts and output tok/s in bench throughput --output-json
- [Kernel] Add distributed Triton autotune foundation
- [Perf][DCP] DeepSeek-V3.2: skip other ranks' top-k slots in FP8 sparse attention on Hopper
- [Perf][MoE] Prefer Triton for block FP8 MoE on SM120
- [Roadmap] Q4 2026 vLLM × RL
- [Weight Reload][GLM5-next] Refresh derived caches in stable storage
- Docs
- Python not yet supported