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] dp: Add shape-aware DP load balancing method
- [Fix] moe: Support more than 1024 expert buckets in the AOT moe_align_block_size
- [Test] diffusion: Observe actual SDPA backend dispatch
- [Test] diffusion: Opt the custom VAE fixture into trusted code
- [Feat] api-codegen: Tagged payload presence rules and multi-arm untagged lists
- [Feat] rust-server: Pause and continue generation, freeze_gc, ping and ready, and the frontend state
- [Feat] rust-server: set_internal_state, profiling, expert distribution, slow_down and memory occupation controls
- [Feat] rust-server: pd_role_switch and update_weight_version over the model-facts cache
- [Feat] rust-server: Weight reads, update groups, weight updates and LoRA refusals
- [Feat] rust-server: /parse_function_call and /separate_reasoning
- Docs
- Python not yet supported