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
- [Bugfix] DeepSeek-V4 parser: drop the content/tool separator so parsed turns re-render byte-identically
- [Bugfix][Engram] Reject offloaded tables larger than host memory
- [Bug]: GLM required tool-call grammar accepts split marker tokens that the streaming parser treats as content
- [Bug]: [Performance]: Realtime (streaming-input) sessions get ~3 ms slower per step for every minute of audio — O(n) scheduler work in strip_covered_mm_data
- [Bugfix] Report a usage error for a trailing dotted argument with no value
- [Misc][ROCm] Add ROCm support to generate_cmake_presets.py
- [Bugfix][Frontend] Preserve MiniMax M2's unclosed final tool parameter
- [Bug]: Qwen2-VL video sampling divides by zero after rounding a one-frame clip to zero frames
- [CI][Test] Retry RemoteOpenAIServer startup on early death with a fresh port
- [Bug][XPU][MRV2]: engine dies when the first request after startup is a large prefill (eager all-reduce after XPU graph capture leaves the stream "Recording")
- Docs
- Python not yet supported