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
- [Bug] --strip-thinking-cache + retraction: release_kv_cache frees KV slots the radix tree still owns (double free)
- sglang-mm: add InternVL server-pipeline processor
- feat(amd): add native GLM MXFP4 Triton Gluon MoE backend
- [NPU] Transfer MTP draft pools in MLA kernel_ascend L2 IO
- [DSV4.1] Remove host syncs when building late-layer tail metadata
- [NPU] add e2e test for some parameters on NPU
- [KV Cache] Amortize L1 eviction under sustained pressure
- [AMD] Limit AITER DSA decode metadata to the active batch
- [AMD] Reuse AITER DSA decode metadata across layers
- [AMD] Fuse AITER DSA decode RoPE and KV-cache preparation
- Docs
- Python not yet supported