CogVideoX-Fun-V1.5-5b-InP-mlx-q8 / quantize_config.json
dgrauet's picture
Upload MLX model via mlx-forge
8599bd4 verified
raw
history blame contribute delete
129 Bytes
{
"quantization": {
"bits": 8,
"group_size": 64,
"skip_components": [
"text_encoder",
"vae"
]
}
}