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
- [Simulator] OFFLINE mode includes predictor query time in cpu_overhead
- feat(grpc): support bounded profiling with num_steps
- fix(kv-events): support per-rank IPC publisher endpoints
- perf: compile grammars and apply vocab mask only on the entry TP rank
- Add DeepSeek-V4 decode radix cache with hierarchical reuse
- [Bugfix] Fix SM120 DeepGEMM grouped FP8 weight requantization
- [Simulator] Exclude predictor query time from offline latency
- [Bug] FLUX.2 rejects --attention-backend sage_attn: model-level whitelist excludes SAGE_ATTN (regression since #22423)
- [ROCm] Route per-head QK norm to the stride-aware JIT kernel instead of copying
- [Bug] Fatal multimodal processor lookup errors omit processor import failures
- Docs
- Python not yet supported