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] sampling: Make penalizer execution order deterministic
- [Fix] lora: Skip unsupported embeddings when targeting all modules
- [Fix] benchmark: Support Gemma 4 in the CSGMV LoRA tuner
- [Fix] [AMD] modelopt: Support ModelOpt FP8 Nemotron models on gfx950
- [Perf] [AMD] spec: Skip the vocabulary-wide sort in MTP rejection-sampling verify
- [Perf] [AMD] dsv4: Fuse the next RMSNorm into mhc_post_pre
- [Test] [NPU] HiCache: Reduce host memory allocation
- [Fix] [NPU] Support bounded encoder-decoder SDPA graphs
- [Fix] modelopt-nvfp4: Resolve the automatic MoE runner once so the weight prep matches the kernel path
- [Fix] mem_cache: Agree on one TreeCore backend across ranks sharing a prefix tree
- Docs
- Python not yet supported