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 supported26 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix] Fix Triton compile crash in penalties kernel on sm_89
- [MM][Perf][CG] Support ViT full cudagraphs for mllama4
- [Core][KV] Retain prefix-cache across hybrid SWA+Full via `is_pinned` blocks
- [Aiter][ROCm] gdn_linear_attn kernel fusion
- [Feature] Triton INT4 / INT2 per-token-head KV cache quantization
- [Reasoning] Add GLMReasoningParser with embedded tool-call rescue
- [Bug]: For Qwen3.5 serise, Large benchmark gap (~10 points) between offline vLLM inference and vLLM API server despite aligned input_ids
- [Bugfix] Fix load balancer waiting count
- [Kernel] Multi-LoRA MoE CUDA BGMV kernel
- [vLLM IR] Support optional output tensor in dynamic_group_quant_fp8 and refactor legacy calls
- Docs
- Python not yet supported