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
- [Bugfix][Kernel] Fix two block-FP8 linear crashes on SM12x
- [BI][DSv4.1] Make FlashMLA sparse attention and the indexer batch invariant
- [Bugfix][Test] Qwen4Exp QSA: run the fp8 kernel tests the way the layer does below sm_89
- [BI][DSv4.1] Make SWA bounded replay exact under batch invariance
- [BI] Make DeepSeek MoE and the EP exchange batch invariant
- [BI][DSv4.1] Run mHC on fixed-split TileLang kernels under batch invariance
- [Bug]: CPU WNA16 MoE cannot load poolside/Laguna-XS-2.1-INT4
- [Perf][Quantization] Honor FP8 activations for Humming linear on SM100/SM110
- [ROCm][Perf][Qwen3.8] Dense mono decode on 128-CU gfx950 (MI350P)
- [Bugfix][EC] Reuse runner cache hits before remote admission
- Docs
- Python not yet supported