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 supported35 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- Optimize ppc64le vLLM runtime image size
- [Performance]: Triton attention drops long-context decode from 3D split-KV to 2D at batch ≥ ~12, ~doubling ITL
- [Usage]: Does it support the deployment of P node TEP and D node DEP modes
- [Frontend] Support min_p in the Responses API
- [Bug fix] Always pass launch_pdl kwarg in fused_inv_rope_fp8_quant (DeepSeek-V4)
- [Render][1/n] Paged shared memory storage for multimodal.
- [Feature][ROCm]: Dual-stream decode (`MULTI_STREAM_OVERLAPPED` shared experts)
- [Bugfix] Escape control characters in xgrammar choice grammar
- Support online C128 compression for DeepSeek V4
- [Feature]: Support for jina-embeddings-v2-small-en
- Docs
- Python not yet supported