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] [AMD] diffusion: Fuse the FLUX.2 adaLN modulate and drop the Ulysses QKV copies
- [Perf] muse-spark: Overlap non-NVFP4 pre-expert norm with routing
- [Fix] unified-memory: Realize the reclaim certificate that relief opens on the tri-pool
- [Feat] cp: Radix cache for padded CP KV layouts
- [Feat] cp: Hierarchical cache for padded CP KV layouts
- [Perf] entrypoints: Avoid O(N^2) request body chunk concatenation in RequestDecompressionMiddleware
- [Feat] embedding: Add Cohere-compatible /v2/embed API
- [Fix] function_call: Parse null and exponent numbers in Gemma4 tool calls
- [Perf] diffusion: Parallelize heavy component loading
- [Perf] [NVIDIA] eplb: Rebalance the DSpark draft MoE as its own placement family
- Docs
- Python not yet supported