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(epd): fail fast on bootstrap bind errors
- [Bug] A non-object entry in a tool-call array discards the valid calls beside it
- fix(function_call): skip non-object entries in parse_base_json
- fix: drop in-flight abort token from visible output in chunked prefill (#34112)
- [CI] Clean up HiCache file-backend pages after tests
- [Fix] Derive a safe simulated acceptance token
- [ROCm] gpt-oss: slice pre-padded hidden for router/topk in moe_impl
- [ROCm] Unlock the fused RMSNorm+MoE-pad for Quark W4A8 on gpt-oss
- [ROCm] aiter backend: route sinks-model extends through unified_attention (gpt-oss prefill −18…−32% TTFT)
- [ROCm] aiter backend: CK varlen route for long full-attention sinks extends (−15…−16% TTFT at 64k uncached)
- Docs
- Python not yet supported