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 supported59 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: With --enable-fault-tolerance, an async EPLB worker exception stalls EPLB silently instead of making the engine UNHEALTHY
- [Bugfix][Reasoning] Granite: report the reasoning end so structured output applies
- [Frontend] Rust gRPC: support bounded per-request multimodal kwargs
- [Test][Rust] Select engine 0 in sampling metrics test
- [Draft][MoE][NVIDIA] Support MXFP4 experts on SM107 in flashinfer_moe_ep_cutedsl
- [ROCm][Perf] Skip the z copy in the Qwen GDN gated RMSNorm
- [RFC][ROCm]: Batch-invariant kernels for Kimi-K3.1 on MI355X (KDA, DSA/indexer, MXFP4 MoE, LoRA, all-reduce)
- [Test][KV Events] Check a reference consumer against a real engine
- fix(tool_parsers): support Python reserved keywords as tool arguments in pythonic parsers
- [Spec Decode][PP] Centralize auxiliary hidden-state relay
- Docs
- Python not yet supported