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
- [OpenAI] Avoid encode/decode round-trip and id-space splice in jinja path
- Add Gemini review guidance for component rules
- [ROCm Test-First CI]: Day 0 SGLang MI455X UALoE72 Helios Rack open source community upstream CI by AMD Advancing AI July 22 2026
- [sglang-miles] Cherry-pick RL/VLM/Qwen3-MoE fixes
- [WIP] Improve CUDA IPC to make it production-ready
- PD MHA KV transfer may mis-detect draft KV layout for uneven PP full-attn layers
- DeepSeek-V4 support disaggregation-decode-enable-radix-cache and support MTP
- [Parity with CUDA H200] MI300/MI325 DeepSeekv4 Pro Support
- [Fix] Register Qwen3_5ForCausalLM/Qwen3_5MoeForCausalLM for text-only checkpoints
- [Spec] Support topk > 1 (EAGLE tree) in the trtllm_mha attention backend
- Docs
- Python not yet supported