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
- [Perf] mem_cache: Anchor session refs at the prompt end, with a configurable session header (opt-in)
- [Perf] DFLASH: Keep an all-sliding-window draft's KV on the target's SWA slots (opt-in)
- [Fix] diffusion: NDRotaryEmbedding theta_rescale_factor and repeat_interleave_real grid path
- [Feat] anthropic: forward inline MP4 user content to Chat
- [Fix] constrained: Allow uniqueItems false in schema prevalidation
- [Fix] multimodal: Decode JPEGs on GPU with torchcodec and apply EXIF orientation
- [Fix] QSA: Stop speculative verify from overwriting the pending index-key ring
- [Fix] diffusion: LoRA rank-0 download hang, QKV LoRA slicing, IPC sync state
- [RFC] diffusion: LTX-2 stage-1 distilled LoRA base restore is not exact
- [Fix] hicache: Reject incomplete NIXL cache files before reading
- Docs
- Python not yet supported