sglang
https://github.com/sgl-project/sglang
Python
SGLang is a high-performance serving framework for large language models and multimodal models.
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 supported1 Subscribers
Add a CodeTriage badge to sglang
Help out
- Issues
- fix(scheduler): skip user stop checks during the reasoning phase
- fix(anthropic): report stop_sequence when user stop sequences match
- fix(reasoning): avoid empty content when reasoning is truncated by a stop
- fix(protocol): honor top-level enable_thinking field
- Fix: preserve priority when splitting a batched EmbeddingReqInput (#32844)
- [Bug] --log-requests max_length truncation is bypassed for lists: elided long lists skip per-element truncation (JSON), text format never truncates list elements
- Remove redundant call to get_available_gpu_memory
- [Bugfix] Pad DSA topk_indices_offset under prefill CP
- fix(openai): preserve highest-ranked candidate on top_logprobs text cā¦
- BUG:skip when spech4 deepseekv4
- Docs
- Python not yet supported