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 supported43 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: Running DeepSeek-V4 fails with CUDA error: unspecified launch failure in synchronize_input_prep
- [Bug]: DeepSeek V4 pro can not run with TP16
- [Bugfix][Frontend] Return streaming errors before the first token
- [Bug]: vllm does not expose /v1/audio/transcriptions for google/gemma-4-E4B-it
- [Bugfix][ROCm][AITER MLA] Size persistent metadata buffers off max(max_num_seqs, max_cudagraph_capture_size)
- [Tracking] DeepGEMM SM 12.x kernel coverage gaps for DeepSeek-V4-Flash on consumer Blackwell (RTX 50 / GB10)
- [Bug]: Report "System message must be at the beginning." When using qwen3.6-27B
- [EPLB] Add offline expert mapping support (aka "offline EPLB")
- [ROCm][CI] Expand core kernel coverage
- [Bug]: TP=2 deadlock on dual AMD R9700 (gfx1201/RDNA4) — GPUs spin at 100%, inference blocked
- Docs
- Python not yet supported