MochiDiffusion/MochiDiffusion

Run Stable Diffusion on Mac natively

What it solves

Mochi Diffusion is a native macOS app that allows users to run Stable Diffusion and FLUX.2 Klein models locally and offline. It focuses on maximizing performance and reducing memory usage on Apple Silicon Macs, solving the problem of running heavy image generation models on consumer hardware without relying on the cloud.

How it works

The app leverages Apple's Core ML implementation of Stable Diffusion and the Iris pipeline for FLUX.2 (Klein). By utilizing the Neural Engine, it can generate images with high memory efficiency (using as little as 150MB). Users can choose between compute units—either CPU & Neural Engine for a balance of speed and low memory, or CPU & GPU for potentially higher speed on high-end M-series chips.

Who it’s for

It is designed for Mac users with Apple Silicon (M1 and later) running macOS 15.6 or later who want to generate images locally for privacy and speed.

Highlights

  • Local Generation: All processing happens on-device, ensuring complete privacy.
  • Broad Model Support: Supports Stable Diffusion and FLUX.2 Klein models.
  • Advanced Features: Includes Image2Image and ControlNet support.
  • Native Experience: Built with SwiftUI for a native macOS feel.
  • Integrated Gallery: Built-in gallery with import, save, and sync capabilities.
  • Metadata Preservation: Saves prompt information directly into the image EXIF metadata.

Related

  • Dispatch
    Swift Diffusers for Mac 1.1 Release

    Hugging Face has released Diffusers for Mac version 1.1, a native open-source app that leverages Core ML to accelerate Stable Diffusion on Apple Silicon, offering up to 2x faster generation on certain hardware configurations.

  • Project
    easydiffusion/easydiffusion

    A one-click installer and user interface for Stable Diffusion that allows users to run powerful image generation AI locally without requiring technical knowledge.

  • Project
    jjang-ai/mlxstudio

    A native macOS app for Apple Silicon that provides a user-friendly interface to run LLMs, vision models, and image generation locally using the MLX framework.

  • Project
    ShiftHackZ/Stable-Diffusion-KMP

    An open-source, cross-platform AI image generation client for Android and iOS that connects to self-hosted Stable Diffusion servers, cloud APIs, and local on-device runtimes.

  • Dispatch
    Stable Diffusion Core ML on Apple Silicon – How to Run and Optimize

    Hugging Face released Core ML‑converted Stable Diffusion checkpoints for Apple Silicon, enabling on‑device image generation in Python or Swift with up to 18 seconds per image on an M1 Max.