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] Remove _swa_scatter_kernel in target verify in DSV4
- [Bug][Spec] Keep adaptive target graph construction outside draft contexts
- [LoRA] Load Gemma 4 PEFT adapters on attention_k_eq_v checkpoints (gemma-4-31B-it)
- fix(anthropic): preserve reasoning history for Jinja templates
- [Bug] Gemma 4 31B (attention_k_eq_v) PEFT LoRA adapter fails to load: missing v_proj, wrong per-layer buffer dims, vision-tower tensors collide
- [Bugfix][Anthropic] Pass prior-turn thinking as reasoning_content when the template reads it
- [GLM47] Reduce full-assistant grammar overhead: batched masks, thinking fast path, shared metadata
- [Bug] Large HiCache UCX startup timeout on downstream build; not reproduced on v0.5.20
- [Fix] MiMo-V2: accept checkpoints with the split attention layout
- [Fix] Reject an invalid stop_regex with a 400 instead of a 500
- Docs
- Python not yet supported