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
- [AMD] MiniMax-M3: fuse sparse QK norm, RoPE and cache writes with AITER
- fix: guard top-logprobs and token-ids logprob rows against missing data
- feat(cake_gemm): add Cake batch DeepGEMM FP8 backend
- [Kernel] Add H200 Nemotron-3.5 MoE configs
- [RFC] Orchestrator-facing Elastic EP lifecycle contract
- [Fix] DCP: reject fa3 for MLA and stop defaulting to it on Hopper
- Update DeepSeek V4 for B200 FP4 AgentX HiCache MTP
- [Bug] deepep FP8 MoE on GB300/sm_103 crashes in DeepGEMM m_grouped_fp8_fp4_gemm_nt_contiguous — UE8M0 scale-format mismatch (CUDA 719 / smxx_layout.cuh:131 assert)
- [Kernel] Tune Kimi-K3 Marlin MoE block sizes on H200
- model_loader: only open the MTP-head shards for a NextN/MTP draft load
- Docs
- Python not yet supported