Modelpublic

yujiepan/dreamshaper-8-lcm-openvino-w8a8

sourceHugging Facecreativeml-openrail-mupdated 11mo agoView on Hugging Face

1likes24downloads

Model Card

yujiepan/dreamshaper-8-lcm-openvino-w8a8

This model:

1.applies latent-consistency/lcm-lora-sdv1-5 on base model Lykon/dreamshaper-8;
2.conducts W8A8 post-training quantization on unet model;
3.converts other models (text encoder & vae) in FP16;
4.and is finally converted as OpenVINO model.

Usage

python

from optimum.intel.openvino.modeling_diffusion import OVStableDiffusionPipeline
pipeline = OVStableDiffusionPipeline.from_pretrained(
    'yujiepan/dreamshaper-8-lcm-openvino-w8a8',
    device='CPU',
)
prompt = 'cute dog typing at a laptop, 4k, details'
images = pipeline(prompt=prompt, num_inference_steps=8, guidance_scale=1.0).images

[image]