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
- [Docs] Fix JIT kernel example paths
- fix: exclude NaN-producing autotune config for DSA Triton prefill on ROCm
- fix: unblock the DSA prefill split-op graph surface on HIP (is_cuda gate)
- [Spec] Stage DRAFT_EXTEND_V2 and point per-step fence at verify record
- [Bug] MLA + fp8 KV: fa3 and flashinfer_mla dequantise the ENTIRE per-layer KV pool every layer, every forward (43.7% of a decode step on DeepSeek-R1 / 8xH200)
- [Bug] Streaming ASR emits partial-word suffixes at word revisions
- [Bugfix] Preserve word boundaries in streaming ASR rollback
- [Bugfix] Guard harmony parsing against malformed gpt-oss output in Responses API
- [Speculative] Enable target-only mixed chunked prefill for DSpark
- https://github.com/sgl-project/sglang/issues/10038[Bug]
- Docs
- Python not yet supported