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
- [Perf] pd: Copy the last chunk's state page indices to the host without a scheduler sync
- [CI] [AMD] workflows: Fix coverage gaps, gating and Extra reporting
- [Perf] [AMD] dsa: Replicate the DCP query projection on the owned route
- [Fix] [AMD] deepseek_v2: Avoid PPMissingLayer attribute errors in the gfx950 allocator path
- [Fix] diffusion: Preserve FP32 product rounding in Anima RoPE
- [Fix][CPU] Handle unaligned GPTQ INT4 output widths
- [Fix] [AMD] dsv41: Refuse PD peers that differ in the page-8 index-K layout
- [Perf] [AMD] attention: Switch MLA decode to Aiter on gfx1250
- [Fix] [AMD] DSv4.1: Use 64-bit row offsets in the candidate-block kernels
- [Feat] tool-call: --tool-call-constraint none samples auto tool choice without a grammar
- Docs
- Python not yet supported