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 supported46 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- fix: simplify dependency graph for dependents of LoRA utilities
- [Bugfix] Zero recycled KV cache blocks for FullAttention models
- Add fused MoE LoRA kernel tuner (tune_lora_moe.py)
- [Perf] Async GPU P2P access cache precomputation to reduce startup time
- Support NVFP4 per-expert weight loading for Gemma 4 MoE
- [Feature]: Support routed_experts export in disaggregated Prefill/Decode serving
- feat: skip-softmax support for FlashInfer attention path
- Fix fp8_e5m2 KV cache blocked for AWQ/GPTQ models
- [Model Runner V2] Skip draft propose for disagg prefill instance
- [Bugfix] Fix Voxtral Realtime crash on mixed-batch transcription
- Docs
- Python not yet supported