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
- [Inkling] Bound audio embedding memory
- Emit dense LM-head logits directly in FP32 on CUDA
- CI: launch test servers in their own sessions and kill by process group
- fix: preserve speculative draft weights across release/resume_memory_occupation
- feat(gateway): support /abort_request in sgl-model-gateway
- [kv-canary] Scatter req token ids by (request × column-block) instead of per-token owner search
- Allow runtime schedule_policy switching via /set_internal_state
- [AMD][Deepseek-R1] Switch attention output projection mxfp4 gemm from triton -> asm
- Dcp with hicache
- [Bug] DSpark speculative decoding causes accuracy regression on DeepSeek-V4-Flash (AIME25 97.08→93.96)
- Docs
- Python not yet supported