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
- [Frontend] Add completion SSE on top of packed top-k logprobs
- [Bugfix][Core] Reclaim parked streaming sessions in reset_prefix_cache
- [ROCm] ROCm binding on common Kimi-K3 Dspark MLA module
- [ROCm] Enable tuned AITER per-token FP8 linear on RDNA4 (a8w8 on RDNA3 WIP)
- [Perf][MoE] Wrap GateLinear ll_bf16_gemm in custom op to prevent torch.compile branch freezing
- [Feature] Red-Green watermarking
- [Bug] pause_generation(mode="keep", clear_cache=True) raises when a streaming-input session is parked between turns
- [Bug]: grouped_topk selects wrong experts with more than eight groups
- [Bug]: Qwen3.5 (hybrid GDN) models cannot run dflash speculative decoding with pipeline parallelism (missing supports_aux_hidden_states_over_pp opt-in)
- [Bugfix] Use native memory fence for SHM broadcasts
- Docs
- Python not yet supported