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
- [Kernel] Fuse native BF16 ReLU2 with NVFP4 quantization
- [Bugfix] Return real upstream errors and evict failed decoders in the XpYd proxy demos
- [Bugfix][Executor] Size response rings for in-flight batches
- [CPU] Enable INT4 and grouped-gemm MoE kernels on AVX-512
- [Security] Clamp Step3-VL images before square-pad allocation
- [BugFix][CPU] Pin CPU builds of torchaudio, torchcodec and torchvision on Linux
- [Bug] Marlin int8-activation path reads negative group scales as unsigned, corrupting every such group
- [Bugfix][LoRA][MoE] Use int64 token offsets in MoE-LoRA Triton kernels
- [Bug]: V2 runner handles KV preemptions after request updates can overwrite pages
- [Bug]: Qwen3.6-35B-A3B-FP8 model with TP 4 gives gibberish output !!!!!!!!!!!! on Intel B70 cards
- Docs
- Python not yet supported