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
- [RFC] First-Class, Versioned KV Hint Envelope for SGLang
- [Performance] Multiple potential avoidable scheduler, sampling, and streaming overheads identified
- [Fix] Reasoning parser: bare </think> from hybrid thinking models leaks into content
- [DeepSeek-V4][PD] Enable CP-v2 interleave for FlashMLA sparse prefill
- [Diffusion] Add MiniMax-H3 packed varlen path for sla_attn backend
- test: add KL guards for HiCache buffer-only host memory mode
- [Bug] Fix HiCache storage keys to include kv_cache_dtype — prevent silent cross-run cache collisions
- [diffusion] acceleration: add SeaCache algorithm
- Observability and Logging capabilities enhancement for HiCache
- [Bug] Streaming session client disconnect can silently drop session context and crash the scheduler
- Docs
- Python not yet supported