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
- [Playground] Verified cell: h100 / default / fp8 / low-latency / single
- [Fix] Preserve GLM4 comma argument text instead of Python tuples
- [Spec][DSV4] Fix dense DSpark draft KV pool under CP: shard by the draft's TP width and reserve its real cost
- [Fix] Render LLAMA2 message roles and separators from stored roles
- Update Qwen3.8-Flash-Next for GB300 NVFP4
- [Fix] GLM-4.7/5 tool parser: stop turning comma-containing values into arrays
- [DSA][KDA] Fix MoE padding handling in hybrid MTP verification
- [Perf] trtllm_mla: pass host q/kv lengths to the ragged prefill to avoid a device sync
- [Perf] Check tool parameter schemas with jsonschema-rs
- [Fix] GLM-4.7/5 tool parser: stream non-JSON array/object/boolean values as valid JSON
- Docs
- Python not yet supported