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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Bugfix] Fix moe_wna16 scale/qzeros shape when dim not divisible by gā¦
- [Bug]: Missing `reasoning` on Some Turns in Multi-Turn Tool-Calling (DeepSeek-V4-Flash-0731)
- [Refactor] move moe sequence parallel boundaries into parallel linear layers
- [Bug]: No viable structured-outputs backend for Kimi-K2.6 + EAGLE3 on v0.26.0: xgrammar bitmask desync kills the engine, outlines compile-timeout leaks threads until OOM, guidance rejects the slow tokenizer
- docs/design/metrics.md: be consistent about _total suffix
- [KV Connector][Offloading] Certify per-token-head KV quant in the canonical layout
- [Bug]: Inkling-Small structured-output request returns 500 ("grammar rejected tokens [200028]")
- [Bug]: xlam tool parser: streamed tool-call arguments duplicated by end-of-stream flush; markers/JSON leak into content
- [Bug]: gigachat3 tool parser leaks special tokens into streamed content/arguments when they are split across deltas
- [Test] Add a coverage ratchet for per-parser tool-calling tests
- Docs
- Python not yet supported