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] Illegal memory access in online-compress prefill plan and EP scatter kernels
- feat: add Xing4_0 model support
- [Fix]Account for DCP replication when sizing the EAGLE draft KV pool
- Use the 128-row query tile for head_dim <= 128 on gfx950
- feat(hicache): opt-in exclusive L2->L3 tiering, dedup cache-mode L3 writes through the existence cache
- [Bug] Potential DSpark startup hang from an unsynchronized CUDA graph memory check
- test(npu): add e2e smoke test for --radix-eviction-policy-config
- fix(retract): restart output cursors when input_embeds generation restarts
- Fix/compressed tensors nvfp4 hopper
- perf(xpu): promote MXFP4 MoE bias to fp32 at load, not per forward
- Docs
- Python not yet supported