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 supported62 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [KV Transfer] Connector-agnostic KV checksums for P/D disaggregation
- [ROCm][AITER] Enable AITER by default on CDNA3+, keep AMD CI on the non-AITER path
- [Compressed-Tensors] [REAP] Non-uniform expert sparsity
- [ROCm][Step5] Enable shared Step5 Triton kernels
- [Bug]: --reasoning-parser qwen3: leading \n\n left in content after stripping the <think> block
- [Bugfix][Kernel] Preserve output shape in FlashInfer FP8 linear
- [Bugfix][Distributed] Use a file rendezvous for offline single-node DP
- [Tracking][Help Wanted] TP-agnostic KV offloading
- [Docs][Model] Sync supported-model tables with the registry and add a consistency test
- [Bugfix][Frontend] Do not count the token sampled for echo with max_tokens=0
- Docs
- Python not yet supported