How to use zjyao-PKU/Pref-Restore-PhaseA-Fidelity with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-to-image", model="zjyao-PKU/Pref-Restore-PhaseA-Fidelity")
# Load model directly from transformers import AutoTokenizer, blip3oQwenForCausalLMVAE tokenizer = AutoTokenizer.from_pretrained("zjyao-PKU/Pref-Restore-PhaseA-Fidelity") model = blip3oQwenForCausalLMVAE.from_pretrained("zjyao-PKU/Pref-Restore-PhaseA-Fidelity", device_map="auto")
How to use zjyao-PKU/Pref-Restore-PhaseA-Fidelity with Sana:
# Load the model and infer image from text import torch from app.sana_pipeline import SanaPipeline from torchvision.utils import save_image sana = SanaPipeline("configs/sana_config/1024ms/Sana_1600M_img1024.yaml") sana.from_pretrained("hf://zjyao-PKU/Pref-Restore-PhaseA-Fidelity") image = sana( prompt='a cyberpunk cat with a neon sign that says "Sana"', height=1024, width=1024, guidance_scale=5.0, pag_guidance_scale=2.0, num_inference_steps=18, )
