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
- [KV Cache] Use exact page demand for paged prefill eviction
- [Diffusion] Fail MiniMax-H3 quality="high" closed on substituted DiT weights
- [Bug] The timeout setting is invalid in can_terminate_prefetch when pool_transfers_done is false
- Reject invalid base64 image payloads
- [diffusion] Add hierarchical startup-time breakdown (SGLANG_DIFFUSION_STARTUP_PROFILE)
- [AMD] Add ROCm 10 (gfx942 / gfx950) stages to the ROCm Dockerfile
- [CI] Add dinesh20073 to CI_PERMISSIONS.json
- [Bug] NIXL/UCX prefill segfault (nixlUcxSharedThread -> cuEventQuery) is not fixed — reproduces on v0.5.17 / CUDA 13.0 / B200; #23489 and #23499 were closed without a root cause
- [Docs] Fix stale env-var defaults: SGLANG_SANITIZE_NAN_LOGITS, SGLANG_MORI_*, SWA eviction interval; drop 2 removed vars
- [diffusion] Allow per-role attention backend override
- Docs
- Python not yet supported