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
- [Bug] DeepSeek-V4 long-context prefill: illegal memory access in the DSA indexer top-k kernel (`topk_v1.cuh:348`); the paged prefill path can never reach the v2 kernel
- fix(topk): preserve candidates across radix buffer overflow
- [Bugfix] Neutralize literal GLM-V media placeholders
- fix(quantization): per-stream Marlin workspaces to prevent cross-stream CTA deadlock under PD-Multiplexing
- Reproducible EAGLE `vectorized_gather_kernel` OOB with concurrent constrained tool request → NCCL abort / Xid 43
- [Fix] Skip the chunk-boundary mamba checkpoint when no slot can be donated
- [Spec] Move state capture outputs with the accepted path (EAGLE topk>1, ngram)
- Make prefill shared-read completion explicit and default-on
- [Bug] VisionTritonAttention drops MiMo-style window attention and sinks
- [Bug] Responses cancel endpoint accepts non-background responses
- Docs
- Python not yet supported