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
- [Test] [NPU] kimi_k3: enable L2 host HiCache for 32p gpqa accuracy test with host hit-rate bench
- [Feat] gemma4: per-layer attention backend
- [Feat][XPU] minimax-h3: add xpu api for indexed_scale_shift_bf16_ and indexed_gate_bf16_
- [Fix] step5: Keep FA4 SWA graph buffers fixed and accept streamed weight updates
- [Fix] [AMD] GLM-5.3-Flash: Keep the MoE correction bias fp32
- [Perf] [AMD] GLM-5.3-Flash: One-launch MoE router and narrower KDA V tile for MTP verify
- [Perf] [AMD] GLM-5.3-Flash: Fewer DSA k-pool indexer launches in MTP verify
- [Fix] Fix bootstrap routing for plain DP replicas
- [Fix] hicache: tolerate concurrent file storage directory creation
- [Fix] qwen3.5: Preserve encoder weights under pipeline parallelism
- Docs
- Python not yet supported