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
- Add GELU activation function to dispatch to existing TRTLLM NvFP4 MoE backend
- [Build] Fall back to system libgomp when torch has no vendored copy
- [MM][Perf][CG] Support ViT full CUDA graph for glm4_1v image and video inference
- Memshare
- [Bugfix] Gemma-4: Add bnb QuantState alias hook on k_eq_v to load Gemma-4 BNB 4-bit weights
- Revert "[MoE Refactor] Add more MoE layer tests" (#39349)
- [Kernel][MoE] Add A100 tuned config for E=64,N=1408 (Kimi-VL-A3B, GLM-4.5 Air)
- [Feature]: Dynamic PDL Enablement
- [Perf] Gemma4: output int64 from fused routing kernel to avoid redundant dtype copy
- [Bugfix] Skip tool parsing when no tools are provided, add unified Mistral parser
- Docs
- Python not yet supported