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] Hard watchdog: log and send SIGQUIT even if diagnostics hang
- [Fix] hicache/pd: Promise prefill only the L3 span the decode prefetch takes
- [Fix] Preserve inline system messages for the dsv41 native encoder
- [Bugfix] Reject unsupported mask in /v1/images/edits instead of silently ignoring it
- [Fix] K-EXAONE 1.0: expose the sliding window to the attention backends [1/5]
- [Bugfix][SM120] Requant block-FP8 MoE weights to UE8M0 for the DeepGEMM grouped GEMM
- [PD] Runtime role switch: engine launched with --disaggregation-mode decode crashes with illegal memory access after decode -> prefill -> decode
- [ROCm][Perf] Qwen3.8-27B dense mono decode: persistent FlyDSL GDN and FFN launches (TP1 and TP2, MI355X/MI350X)
- [Fix] mamba: Release consumed lazy Mamba prefix locks before donation retry
- fix(spec): prefer decode and default EAGLE chunked prefill to 1024
- Docs
- Python not yet supported