Instructions to use iky1e/moebius-ft-celebahq-mlx-f16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use iky1e/moebius-ft-celebahq-mlx-f16 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir moebius-ft-celebahq-mlx-f16 iky1e/moebius-ft-celebahq-mlx-f16
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
File size: 617 Bytes
9956540 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 | {
"formatVersion": 1,
"modelName": "ft_celebahq",
"sourceCheckpoint": "Moebius-Models/ft_celebahq/diffusion_pytorch_model.bin",
"precision": "f16",
"imageSize": 512,
"latentSize": 64,
"latentChannels": 4,
"maskChannels": 1,
"conditioningIDs": 20,
"scalingFactor": 0.13025,
"noiseOffset": 0.0357,
"scheduler": {
"betaStart": 0.00085,
"betaEnd": 0.012,
"trainTimesteps": 1000,
"betaSchedule": "scaled_linear",
"clipSample": false
},
"files": {
"unet": "unet.safetensors",
"vaeEncoder": "vae_encoder.safetensors",
"vaeDecoder": "vae_decoder.safetensors"
}
}
|