AI GitHub
Apple Stable Diffusion for Core ML deployment
Convert supported diffusion models and run them through Apple's Core ML stack.
Key Takeaways
- The repository provides conversion and inference examples for supported Stable Diffusion pipelines.
- Plan conversion and application integration together.
- Convert a documented checkpoint and reproduce a small image on the target device.
- A model that runs on a Mac can exceed memory limits on a phone.
How it works
The repository provides conversion and inference examples for supported Stable Diffusion pipelines. Converted components and the application runtime must agree on their interface and resource assumptions.
What you need
Plan conversion and application integration together. The app needs all converted components, preprocessing resources and device-compatible runtime settings.
What to try first
Convert a documented checkpoint and reproduce a small image on the target device. Measure initialization, per-image runtime and peak memory separately.
Test a device with less memory than the development machine. Startup, model loading and image generation can peak at different times.
Where to look for failures
A model that runs on a Mac can exceed memory limits on a phone. Verify device compatibility and artifact size before integrating it into an app.
Related Franklin pages
Match the pipeline and device to a specific model variant. Compare a controlled prompt on known assets before testing a new checkpoint, conditioning input or export.
- Use modular diffusion pipelines for supported image, video and audio models. Diffusers for building generation pipelines in Python
- Guide supported image generation models with inputs such as edges, poses or depth. ControlNet for conditioning diffusion with spatial cues
Franklin’s image tests give you examples of outputs to compare across visual models. Franklin’s image tests
Project documentation
Try it for your workflow
Explore the project’s documentation for setup and usage, or save this page to a collection for later.
Read the project documentation ↗