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] DSv4.1: Gather the compressor projection instead of the hidden states under prefill CP
- [Spec] Fix trtllm_mha draft startup on GLM-5.3-Flash
- [Feat] HiCache: Add PP storage query and completion consensus
- [Perf] function_call: Avoid O(N^2) buffer rescanning in KimiK3Detector streaming parser
- [Perf] function_call: Avoid O(N^2) regex backtracking and buffer reslicing in GLM4/Ling3 streaming parsers
- [RFC] linker: Prefetch SSD-resident KV while the request is queued
- [Fix] Anthropic: accept thinking.display="updates"
- [Fix] openai: Return errors in the format OpenAI clients expect
- [Fix] openai: Report reasoning tokens where OpenAI clients read them
- [Fix] openai: Report aborted requests as an error or a stop
- Docs
- Python not yet supported