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 supported50 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Core][Metrics] Add exact recomputed-token counter
- [Bug]: qwen3_coder tool parser drops the last parameter when the model omits </parameter> before </function> (non-streaming {}, streaming unterminated JSON)
- [Bug][Scheduler][Mamba] Track uncomputed align checkpoint visibility after same-step priority preemption
- [Feature][ROCm][Perf]: Add AITER gluon sparse MLA for rope-free BF16 (GLM-5.3-Flash, gfx950)
- [Feature]: Share non-streaming chat message assembly between `OpenAIServingChat` and batch derender
- [ROCm] Enable Hisparse on ROCm : Sparse MLA Hot-Buffering
- [Docs] Clarify nightly installation and fix cu129 resolving to CUDA 13 wheel (Refs #42338)
- [Build] Cap default build parallelism by available memory in setup.py
- [Bugfix][Core] Release pinned host cache before discarded wake remap
- GLM-OCR MTP broken by two defects: loader drops model.language_model.layers.* MTP weights before prefix rewrite; MTP forward not CUDA-graph safe
- Docs
- Python not yet supported