Ling Cookbook
A collection of deployment recipes and practical guides for running the Ling model family locally with SGLang, vLLM, llama.cpp, Ollama and MLX across consumer and workstation hardware.
Ling Cookbook provides deployment recipes and engineering examples for the Ling model family from inclusionAI, organized under a local-deployment guide for consumer and workstation hardware.
For Ling-3.0-flash, it covers recipes for a single NVIDIA DGX Spark using SGLang or vLLM at several quantization levels, and llama.cpp with GGUF quantization. For Ling-3.0-tiny, it covers SGLang, vLLM and llama.cpp on a DGX Spark, plus Ollama, llama.cpp and MLX on Apple Silicon Macs.
The repository also distributes pre-built agent skills for the models' multimodal and agentic capabilities, which can be downloaded for local use or inspected by an agent that clones the repository to audit their safety before use. It is a work in progress, written mostly as Jupyter notebooks.