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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [MoE/b12x] Auto-select FLASHINFER_B12X NVFP4 MoE backend on SM120, reject EP deployments
- [RFC]: Opt-in incremental prompt encoding for multi-turn chat
- [Bugfix] [EPLB] Avoid Silent Accuracy Corruption in Quark MXFP4 EPLB
- [Performance]: benchmark speed & precision of rust preproc vs pytorch
- [Performance]: Native MTP draft acceptance rate decays with total context length (Qwen3.6-27B, in-range context, no --hf-overrides)
- [Bug] ~17.5s engine-wide stall on large-to-small request shape transitions (NVFP4/b12x, GDN CuteDSL prefill kernel)
- Please ignore. This is to quickly locate CI issue.
- [Bugfix] Require quack-kernels>=0.4.1 for cutlass-dsl 4.5+ compatibility
- [ROCm] Add VLLM_ROCM_ENABLE_CUDAGRAPH forcibly enable cudagraph for encoder-decoder models
- [Kernel] ReplaySSM: cache SSM inputs instead of state for faster standard and speculative decode (Mamba2 + GDN)
- Docs
- Python not yet supported