Sana How to use clark-labs/clark-air-sana-1.6b-mlx-2bit 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://clark-labs/clark-air-sana-1.6b-mlx-2bit")
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,
) 
