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
- [Bug] `pause_generation(abort)` + `release_memory_occupation` right after a client disconnect crashes scheduler
- [Feat] hicache: Add opt-in legacy Mooncake cache fallback
- [Perf] kda: Reuse one int64 cu_seqlens per batch in FlashInfer prefill to avoid per-layer host sync
- [Perf] [Intel] gemma4: Enable frozen-KV MTP graph capture
- [Perf] gemma4: Skip ranking for full MTP chains
- [Feature] Prepare registered radix cache backends before worker graph initialization
- [Feat] mem_cache: prepare registered backends before worker graphs
- [Fix] multimodal: Clear the embedding cache in Scheduler.flush_cache
- [Fix] diffusion: Respect ancestor cgroup memory headroom
- [Fix] [NVIDIA] modelopt: Keep SM12x FP8 batch-invariant in deterministic mode
- Docs
- Python not yet supported