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: complete production metrics reference
- [Bug] v0.5.14: With EAGLE/MTP on, KV pool profiler leaves ~50 GB VRAM idle on hybrid GDN model (Qwen3.6-27B NVFP4)
- Emit HiCache L3 KV events
- For hybrid sliding-window (SWA) models the SWA KV pool is small and quickly
- [Diffusion][CPU] Add GELU kernels and enable RMSNorm CPU path
- Fix duplicate tool_index in PoolsideV1Detector non-streaming path
- [Perf][CP]: optimize glm5.2 prefill performance using comm-comp overlap triton kernels
- [Feature][AMD] Officially support consumer Radeon RDNA3/RDNA4 (gfx1100/gfx1201) — tracking + enablement plan
- [Bug] HiCache prefetch all_reduce deadlock with TP=4, no PP — mismatched call count in check_prefetch_progress
- [PD] Add decode-side bootstrap timeout cleanup for Mooncake
- Docs
- Python not yet supported