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
- [PD] Use one logical-range contract for DSA state indices
- fix(mm): stop passing frame_count_limit positionally into load_video
- [HiCache] Support file-backed L3 offloading for DeepSeek V4 with CP
- [Diffusion] Add API key authentication to HTTP server
- [Diffusion] Map checkpoints read-only where host copies are redundant
- Skip doomed CUDA JIT on pre-Ampere GPUs; align supports_fp8 with CUTLASS
- [Feature] DeepSeek-V4 trtllm attention backend: support PD disaggregation (uniform-FP8 KV layout in the PD handshake)
- [Bug] Multi-part text content is rejected for system and assistant messages
- [Bug] Missing layer prefixes bypass per-module quantization setting
- [Spec][Ngram]: Add global Trie/SAM proposal allocation
- Docs
- Python not yet supported