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
- [Feat] disaggregation: Gate Mooncake session recovery on one real transfer trial
- [Feat] diffusion: Support dynamic LoRA on GGUF transformers
- [Feat] hicache: Register draft Mamba state pool
- [Test] [NPU] rl: Add pause/resume scheduler idempotency and 1-NPU smoke tests
- [Perf] hicache: Skip satisfied Full host eviction heaps
- [Feat] [Intel] sensenova: Run SenseNova-U1.5-8B-MoT on Arc Pro B60
- [Doc] cookbook: Capture prefill CUDA graphs up to 16384 tokens in the GLM-5.3-Flash B300 FP8 Low Latency recipe
- [Feat] scheduler: Hardware governor that scales prefill knobs from GPU telemetry
- Add pinned model-file checksum verification to SGLang Diffusion loading
- [Feat] model-loader: cap the bytes the multithreaded loader keeps in flight
- Docs
- Python not yet supported