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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Attention] Support FlashInfer attention sinks on SM8x
- [Bugfix] Validate headless mode args before building the engine config
- [ROCm] Enable DeepSeek-V3.2 / GLM-5.3 sparse MLA on gfx11
- [Kernel][Attention] Prune Triton tile loops under unclamped mm_prefix (Gemma 3 and other prefix-LM models)
- [Feature][MoE] Add expert-capacity token dropping
- [Kernel][Hardware][NVIDIA] Add tuned block-FP8 W8A8 configs for NVIDIA GB10
- [Core] Add platform-scoped per-request speculative fallback
- [Bugfix][Core] Keep the aligned sliding-window prefix hit when dropping the EAGLE block
- [Bugfix][Reasoning] Step3: scope is_reasoning_end to the current turn
- [Test] Skip FP8 Triton kernel tests on pre-SM89 CUDA GPUs
- Docs
- Python not yet supported