- Home
- AI image generators
- Mochi Diffusion
On the recordRead 12 Oct 202620 facts
Mochi Diffusion
On the record
- Compatibility
- The stated requirements are an Apple Silicon Mac with M1 or later and macOS 15.6 or later.github.com · read 12 Oct 2026
- FLUX.2 Klein inputs
- FLUX.2 Klein can use up to four input images, which the app may resize to fit the attention budget.github.com · read 12 Oct 2026
- FLUX.2 Klein limit
- FLUX.2 Klein generation is limited to four steps and does not use CFG, ControlNet, or negative prompts.github.com · read 12 Oct 2026
- Gallery
- It includes a built-in gallery with image import, save, and sync support.github.com · read 12 Oct 2026
- Hardware and OS
- The project lists Apple Silicon (M1 and later) and macOS 15.6 or later as requirements.github.com · read 12 Oct 2026
- Image metadata
- Generated PNGs include generation settings that AUTOMATIC1111-compatible tools and Civitai uploads can read.github.com · read 12 Oct 2026
- Input images
- FLUX.2 Klein can use up to four input images, which the app may resize to fit pipeline constraints.github.com · read 12 Oct 2026
- License
- The GitHub repository identifies the project license as GPL-3.0.github.com · read 12 Oct 2026
- Local generation
- Image generation happens locally and offline, and the project says it sends nothing to the cloud.github.com · read 12 Oct 2026
- Model limitation
- Each converted Core ML Stable Diffusion model generates images at a fixed size baked in during conversion.github.com · read 12 Oct 2026
- Model requirements
- Using Mochi Diffusion requires Core ML Stable Diffusion or FLUX.2 Klein models.github.com · read 12 Oct 2026
- Model support
- The app supports Core ML Stable Diffusion models and FLUX.2 Klein models; Stable Diffusion models must be converted for Core ML, while Klein models need no conversion.github.com · read 12 Oct 2026
- Offline generation
- The app generates images locally and completely offline, and says that no data or telemetry is collected or sent to the cloud.github.com · read 12 Oct 2026
- Privacy
- The project says Mochi Diffusion collects no data or telemetry.github.com · read 12 Oct 2026
- Purpose
- Mochi Diffusion is a native macOS app for generating images locally with Core ML Stable Diffusion and FLUX.2 Klein models.github.com · read 12 Oct 2026
- Reference images
- Yesgithub.com · read 20 Sep 2026
- Stable Diffusion limit
- Each converted Core ML Stable Diffusion model can generate images only at the fixed size baked into that model.github.com · read 12 Oct 2026
- Support
- The project directs bug reports and feature requests to GitHub issues and code contributions to pull requests.github.com · read 12 Oct 2026
