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
- [Feature] Scheduler: opt-in NO_TOKEN admission lookahead with head-prefix pinning (head-of-line relief for prefill)
- [Bug] ReplaySSM CPU backup/restore loses pending accepted updates
- [Qwen4] Add opt-in structured-output startup warmup
- [Qwen4][SM120] Port opt-in GDN accepted-state recovery (RecoverSSM)
- [Qwen4][SM120] Port opt-in online MXFP8 and rowwise FP8 weights
- fix: preserve pending ReplaySSM updates during CPU retraction
- [Scheduler] Opt-in NO_TOKEN admission lookahead with head-prefix pinning
- fix(attention): unwrap hybrid backend for FP8 DSA MHA
- [4/N] Quantization Refactor: unify the scheme ABCs
- [DSA] Keep superseded trtllm-gen counter buffers alive for captured CUDA graphs
- Docs
- Python not yet supported