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
- [Fix] disaggregation: Preserve early TransferInfo across bootstrap room clear
- [Fix] [AMD] decode-attention: Use two pipeline stages for MLA decode on RDNA3.5 (gfx115x)
- [Fix] constrained: make json_schema_validation subschema traversal deterministic and skip boolean no-op recursion
- [Fix] pplx: Import the FP8 group quantizer from sglang.kernels
- Add DeepSeek V4.1 NPU support and align with upstream changes
- [Fix] model_loader: Load checkpoint weights for every module in layered mode
- [Fix] diffusion: Read encoder folding dims from a nested text_config
- [Perf] diffusion: Keep a folded Gemma-4 encoder's vocab table on the host
- [Perf] [NVIDIA] diffusion: Exchange Q/K/V as E4M3 under Ulysses with fp8_fa_sm120
- [Fix] model_loader: Support BF16 weights in NumPy cache
- Docs
- Python not yet supported