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] scheduler: Add SLA-constrained dynamic batching for decode phase optimization
- [Test] [NPU] npu_memcache: Add L3 hit-rate checks and MetaService bootstrap
- [Feat] [NPU] memory-saver: Release device communication domains on sleep mode
- [Perf] [NVIDIA] quantization: Add tuned channelwise FP8 GEMM tile configs for RTX PRO 5000
- [Perf] deepseek_v4: Score the DeepSeek-V4.1 prefill indexer on Hopper…
- [Fix] [AMD] quark: do not declare the MiniMax-M3 TP8 MoE pad to CK-Tile
- [Fix] multimodal_gen: Use safe_pickle_loads for DiffusionServer inbound requests
- [Feat] hisparse: Share prompt prefixes on the decode host-tier radix cache
- [ROCm][AMD] DeepSeek-V4.1-Flash gfx950 / MI355X Performance Optimization (SGLang)
- [Perf] hicache: Add an opt-in O_DIRECT path to the fast_file backend
- Docs
- Python not yet supported