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
- [Perf] qwen3-vl: Bound vision MLP activation memory for large patch batches
- [Fix] hicache: Drop the warning and comments that blame copy-engine queueing for forward stalls
- [Fix] [CPU] kvcache: Validate indices range in store_cache_cpu
- [Refactor] [NPU] operators: Migrate sgl-kernel-npu calls to public operators entry points
- [Feat] linker: Export load and offload bytes by pool
- [Perf] [AMD] dsv4: Decoder tail graphs and any-context breakable prefill graph for V4.1 on HIP
- [CI] diffusion: Stabilize test execution and resource usage
- [Feat] bench: Record per-request HTTP status and finish reason
- [Perf] [AMD] rotary_embedding: tile fused RoPE and KV write for large T on gfx1250
- [Feature] Optimize Qwen3-VL reranking with concurrent, exact-label, prefill-only scoring
- Docs
- Python not yet supported