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 /server_info 500 error when load_format is set to a custom loader class
- [Platform] Route OOT available-memory queries through current_platform
- fix(config): derive scaling in the KimiVL, MiniCPM3, and DeepseekVL2 MLA branches
- FP4 KV Roadmap
- perf: add dLLM consumer-sufficient TP vocab state
- Auto-select MoE runner backend for DeepSeek V4 FP4 experts
- Fix race on shared ZMQ sockets in mooncake KV manager senders
- [docs] Multi-node deployment: add PD disaggregation and Apptainer examples for SLURM
- Fix device mismatch when mixing JPEG (GPU-decoded) and other type (CPā¦
- [Bug] runai_streamer load-format silently corrupts GLM-5.2 (glm_moe_dsa) weights under TP8 ā loads clean, generates token-0 garbage
- Docs
- Python not yet supported