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
- [KDA] End-to-end context-parallel prefill for hybrid linear models (contiguous strategy)
- managers/detokenizer_manager: fix spurious eviction in LimitedCapacityDict when updating existing key
- Feat/minimax m3 eagle3 dcu
- [Fix] overlap scheduler: record_stream mix_running_indices for the forward-stream relay gather
- [Bugfix] DSA indexer: the Torch reference rejects the shape its caller passes
- [Fix] Guard q_b_proj weight probe so DeepSeek AWQ checkpoints load
- [Bug] Follow-bootstrap-room routing selects inactive reserved DP slots
- fix(gateway): propagate generated request ID to downstream workers
- [Bug] Stock `/abort_request` returns HTTP 200 even though the later abort path fails with `AttributeError`
- docs: python version and environment pinning in MLX docs
- Docs
- Python not yet supported