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
- [Fix] speculative: Pad MLP-sync speculative batches by whole requests
- [Perf] sgl-router: Encode DeepSeek-V4 prompts on the router's fastokens tokenizer
- [Perf] srt: Encode the load snapshot before taking the shared-memory file lock
- [Test] crash-dump: Pin the Python frontend in the tests that resolve --ssl-keyfile-password
- [Perf] hicache: Stage kernel-backend host index uploads in pinned memory
- [Chore] rust-server: Refresh the frontend config catalogue and drop its Rust lexer
- [Feat] kv-shard: Let out-of-tree backends and pools join logical-page KV sharding
- [Fix] [AMD] docker: Pin image source to the checked-out commit
- [Fix] [AMD] config: Reject explicit AITER fusion in deterministic mode
- [CI] [AMD] kimi-k3.1: Skip performance after failed setup
- Docs
- Python not yet supported