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] Use host-backed pool capacity for max_req_len under HiSparse
- [Fix] Spark2.5 MLP use exact GELU to match reference
- [Bug] sgl_kernel ships infllm_ops as a cp310-tagged extension inside an abi3 wheel: unimportable on Python != 3.10, and it carries no sm_121 code
- [sgl-kernel] Add sm_110a and sm_121a gencodes to infllm_ops on aarch64 + CUDA 13
- [Parser] Detect the Spark-X2.5 tool-call parser instead of falling back to glm45
- [Scheduler] Start non-leader ranks from an empty list in the DP-attention request broadcast
- [Fix] Reasoning parser: treat template-opened reasoning that ends on EOS as content (glm45)
- [Bug] glm45 reasoning parser returns the whole answer as reasoning_content (content empty) when the model skips thinking and stops on EOS
- [Bug] --enable-flashinfer-pure-allreduce is not available on 0.5.18
- [Bug][ROCm] GLM-5.3-MXFP4 TP8 + EAGLE GPU memory fault under high-concurrency long-context load
- Docs
- Python not yet supported