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
- [Feature] Selective FP8 KV dequantization for BF16 DSA sparse prefill
- `--api-key`: allow reading the value from an environment variable (avoid exposing it in the process command line)
- [AMD] Enable DSV4 DCP HiCache continuation
- [ROCm] Wave-level optimizations for the top-k transform kernel
- [Bug] NGRAM + `torch_native` is accepted, then crashes during startup warmup on unset `extend_prefix_lens`
- fix: back up chunked prefixes when requests finish
- fix(openai): omit DSV4/DSV32 tools when tool_choice=none
- test(npu): add DSV4-Flash DSPARK / GLM-5.2 / Kimi-K3 gpqa accuracy cases, register to nightly
- [SM120] Enable FlashInfer GDN prefill default for consumer Blackwell
- feat(server): support API keys from environment
- Docs
- Python not yet supported