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 supported61 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Core][Scheduler] Bound the context a prefill step attends over
- [Bug][KV Offload] Reused OffloadKey can lose metadata from delayed BlockStored events
- [Bug]: Native FlashInfer produces garbage output with BF16 Q and non-unit q_scale
- [Bugfix][Multimodal] Fix PaliGemma prefix attention range
- [ROCm][CI] Enable AITER on for Qwen3-8-Flash-Next-FP8 Accuracy Eval
- [Feature]: Worker.sleep device-wide memory check should warn, not assert (false positives with other processes on the GPU)
- [Bugfix][EPLB] Give EPLB each expert-weight region once
- [Core][Spec Decode] Skip draft tokens instead of preempting under KV pressure
- [ROCm][Perf] Muse Spark Open Shard residual stream across TP
- [Bugfix] Warn instead of assert on device memory growth in sleep
- Docs
- Python not yet supported