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
- [NPU] cp-decode-attn-tp: Support dynamic ModelSlim W8A8 weights
- [diffusion] use true_cfg_scale as the FLUX CFG weight
- [diffusion] bound Spectrum under FLUX.1 true CFG and cap CFG gate reuse
- [Fix] Pad the TileLang sparse MLA prefill output back to the padded extend batch
- refactor: TRTLLMHAAttnBackend inherits from AttentionBackend directly
- [Perf] [AMD] qwen3: Fuse QK-norm + RoPE + KV cache write for dense Qwen3 decode
- [Fix] moe: Fix --enable-fused-moe-sum-all-reduce outside CUDA and with LoRA hooks
- [Fix] Require admin API key for weight update endpoints
- [AMD] Fuse MiMo-V2.6 small-M QKV projection on gfx950
- [AMD] Fuse MiMo-V2.6 RoPE and KV-cache insertion on gfx950
- Docs
- Python not yet supported