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
- [Feat] grpc: Add standard health checks to the native server
- [Bug] DeepSeek-V4.1 DSpark with SGLANG_ENABLE_DP_SPEC_PREFILL_COORDINATION: decoder-tail run_idle cuts verify rows to 128 ("next_token_logits row count mismatch")
- [Fix] function_call: preserve split InternLM tool call markers
- [Fix] moe: apply the softmax correction bias on GPU; ERNIE-4.5 router in fp32
- [Fix] glm4_moe: bound the text preamble before a mandatory tool call
- [Fix] [AMD] quark: Preserve W4A4 MXFP4 MoE storage across reloads
- Feat/proxy model gfx1250 1023
- [Perf] [AMD] dsv4.1: Wire up 32x32-block fp8 wo_a GEMM
- [RFC] Activation monitors in SGLang
- [Refactor] multimodal: [sglang-miles] [MM refactor 1/9] Move multimodal enums to a shared module
- Docs
- Python not yet supported