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
Add a CodeTriage badge to vllm
Help out
- Issues
- [Core][Frontend] Add per-request prefix cache telemetry
- [MM] Add a Mooncake-store backend for the multi-modal processor cache
- [Multiproc] Fail fast on concurrent worker exceptions via background error pipe monitor
- fix(warmup): cover all C128A extra_topk widths in sparse-MLA autotune
- [Bug][Spec Decode] DFlash2: draft model unloadable — decoder_layer_cls ignored in DFlashQwen3Model.__init__ + full-vocab init transients OOM 24GB cards
- [Bug]: test_spec_decode_logprobs never runs on NVIDIA CI, and is ill-defined at top-k ties when it does
- [Core][Scheduler] Gracefully handle unexpected request status on KV transfer finish
- Add FlashInfer fused decode for Qwen GDN on sm120
- [Bugfix] Fix DFlash vocabulary module initialization
- EAGLE/MTP prefix-cache last-block drop causes a 1,648-token recompute per hit on one hybrid Qwen3.8 GDN layout — ~30-40% batch throughput loss on prefix-reusing workloads with speculative decoding
- Docs
- Python not yet supported