Verified project record

leejet/stable-diffusion.cpp

A pure C/C++ implementation for running image and video diffusion models locally without Python or heavy dependencies. Users download model weights separately and interact via CLI, web interface, or library integration.

Project overview

Project type: Model Runtime · Image & Vision · Video
Deployment: Refer to project documentation
License: MIT

Best for

Developers who need a local-only, dependency-free library or CLI tool for running diffusion models.
Users who have specific model weights in .ckpt, .safetensors, or .gguf formats and want to execute them without a Python environment.

Key capabilities

Executes generation for image and video diffusion models, including SD, Flux, Wan, and Qwen Image, natively in C/C++.
Provides a built-in web interface for interacting with the diffusion models.
Converts model weights to .gguf or .safetensors formats.

Limitations and risks

Due to active development, the API and command-line options may change frequently.
Telemetry practices are currently not documented.

Getting started

Obtain the sd executable, download diffusion model weights, and run a single command to generate an image. Setup is rated medium difficulty because it requires finding weights and building from source or running commands.

Evidence and sources

README: Plain C/C++ implementation based on [ggml](https://github.com/ggml-org/ggml), working in the same way as [llama.cpp](https://github.com/ggml-org/llama.cpp) - Super lightweight and…
README: ***Note that this project is under active development. \ API and command-line option may change frequently.***
README: - Convert mode supports converting model weights to `.gguf` or `.safetensors`
README: - [Wan2.1/Wan2.2](./docs/wan.md)
README: 🚀 stable-diffusion.cpp now uses a brand-new embedded web UI.

Reviewed at: 2026-07-15T15:50:50Z