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
- [DSV4.1][PD][2/N] Support prefill node use cp parallel strategy + BCG
- [DSV4.1][PD][3/N] Avoid host synchronization in DeepSeek V4.1 CP prefill
- [Unified Tree] refactor: share the aux component eviction driver between SWA and Mamba
- [Perf] Raise RTX 5090 decode CUDA graph default to 128
- [Fix] Honor memory pool debug mode in token allocator
- fix: accept pp_proxy_tensors in frozen-kv-mtp, uno, and multi-layer-eagle forward_batch_generation
- Fix Kimi K2.5 image token accounting in serving benchmarks
- [HiCache] Retry parked cache-mode storage hits on host capacity
- fix(gdn): keep gating beta in fp32 on the extend/prefill path
- [HiCache][perf] Queue and batch write-through backups off the request-finish path
- Docs
- Python not yet supported