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 supported35 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Performance]: Marlin MoE block_size_m heuristic over-selects a large M-tile for expert-parallel decode (up to ~30% TPOT on gpt-oss)
- [Bug]: H200-NVL TP=8 FlashInfer mnnvl fused allreduce crashes at CUDA graph capture; trtllm workaround is ~34x slower than disabling fusion
- [RFC]: Confident Decoding — entropy-guided intermediate-layer logits selection
- [Bug]: Gemma-4 MTP crashes on B200 after FlashInfer selects a missing head_dim=512 TRTLLM-Gen cubin
- [Bug]: DeepSeek V4 reasoning/tool parsers produce unbalanced tags under concurrent load
- [CI Failure]: vllm chat --url : 'url' is deprecated..warning
- [Feature]: Make kv cache dtype releated to platform
- [OOT] Add OOT support for fp8_block linear kernel
- [Bugfix][XPU] Fix Mamba state pointer overflow
- [Structured Output][Tool Calling] Use xgrammar's reasoning parser instead vLLM's
- Docs
- Python not yet supported