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 supported59 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: MoE w2_bias is added tp_size times under tensor parallelism
- [Bugfix][MoE] Count the expert down-projection bias once under TP
- [Bugfix][ROCm][Qwen4Exp] Bind the opaque SP wrappers in the PLE layer
- [WIP][Kernel] Fuse SiLU and FlashInfer SM90 block-FP8 quantization
- [Model] Deprecate PaliGemma support.
- [ROCm][Step5] Use native fnuz FP8 MFMA in CSA prefill indexer scores
- [Bugfix][Frontend] Combine consecutive assistant messages in Anthropic requests
- fix(logits_processor): apply soft capping before scaling in get_top_k…
- [ROCm][Perf][Attention] AITER Gluon paged attention reduce + output gate fusion (Qwen3.8-27B)
- [Perf][GLM-5.3] Fuse MXFP4 MoE input quantization with route. 20% TTFT improvement
- Docs
- Python not yet supported