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
- [DSpark] Support PP verify with stage-local draft owners
- [Disaggregation] Honor admission aborts before KV handoff
- fix(sampling): reject empty stop strings in normalize()
- [router] Honor HF_TOKEN when downloading a gated tokenizer repo
- [Bugfix] Preserve fp32 MoE router precision across model gates
- [diffusion] Enable rank-local TP checkpoint loading for MiniMax-H3
- Bump bundled NVSHMEM to 3.7.2: 3.4.5 lacks NVSHMEM_IB_GID_INDEX, which restricted IB fabrics require
- [Bug] Scripted-runtime rid reuse loses a race with rid_to_state release, surfacing as a 60s recv timeout (3 tests red in test/manual/chunked_prefill)
- [Bug] SWA branching attaches a later Mamba checkpoint to an earlier prefix
- [Bug] Encoder-decoder KV cache: shared boundary page is double-freed when page_size > 1
- Docs
- Python not yet supported