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 supported36 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix][Frontend] Give each batched conversation its own parser
- Add standalone Rust renderer Docker image
- [BugFix] Skip GDN/KDA recurrent state updates for stale zero-accept sā¦
- docs: Add source-level docstrings for NIXL stats and metrics (issue #ā¦
- [Kernel][ROCm] Use naive block assignment for WNA16 MoE
- [Perf] Add fused-MoE tuned configs for NVIDIA GeForce RTX 5090 (E=128, bf16, TP=1/TP=4)
- feat: Unify flashinfer utils into a single file
- [Bugfix][Frontend] Reject unsupported named tool_choice for Gemma 4
- [Perf] Sample in log space in TopKTopPSampler.forward_native, skipping the full-vocab softmax
- [Bug]: [ROCm] DeepSeek-V4 sparse indexer is a dead end on non-MI3xx GPUs, though working torch fallbacks exist in-tree
- Docs
- Python not yet supported