|
--- |
|
license: creativeml-openrail-m |
|
pipeline_tag: text-to-image |
|
--- |
|
|
|
CoreML converted version of Dreamshaper SDXL 1.0 alpha 2 by Lykon (https://huggingface.co/Lykon/DreamShaper / https://huggingface.co/Lykon/DreamShaper/blob/main/ArtDiffusionAlpha2_half.safetensors.safetensors) |
|
Read more about the converted checkpoint here: https://civitai.com/models/4384/dreamshaper |
|
I am not affiliated with the author of the model, all credits go to him. |
|
|
|
Conversion details: |
|
-No quantization |
|
-SPLIT_EINUM model version, it support for Apple's Neural Engine although currently there are some issues with running SDXL models on the Neural Engine, (CPU and) GPU mode is recommended. |
|
-Apple's ml-stable-diffusion package was used for the conversion (https://github.com/apple/ml-stable-diffusion) |
|
-Controlled Unet and regular Unet are both included. |
|
-Compiled & Packaged variants included. |
|
|
|
How to use: |
|
-Can be run using Apple's ml-stable-diffusion package through commandline or with some tinkering via https://github.com/huggingface/swift-coreml-diffusers. |
|
|
|
license (inherited): |
|
-creativeml-openrail-m |
