Just a repo for sharing random CoreML & Stable Diffusion files & info.
October 04, 2023:
This is my current miniconda environment for model conversions.
It is working for SD-1.5 type models, including the ControlledUnet, both original and split-einsum-v1.
It seems faster than the environments that I've run before. Converting ControlNets also works.
I have not tried it with split-einsum-v2, or SDXL.
It uses mostly the current versions of available packages, specifically:
- macOS 14.1
- python==3.9
- coremltools==7.0
- diffusers==0.21.4
- python-coreml-stable-diffusion (pip -e . from a current git clone of ml-stable-diffusio 1.1+)
- safetensors==0.3.3
- scikit-learn==1.1.2
- torch==2.0.0 - - - (torch==2.0.1 also seems ok)
- torch2coreml.py from the current diffusers main repo
The converted models work in both the standard Mochi Diffusion 4.2 and 4.3 and a version I built with Xcode 15 for macOS 14 using a git clone of the current Mochi Diffusion main branch.
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.