Efficient-Large-Model

/

Sana_1600M_512px

like 39

Follow

Efficient-Large-Model 633

not run nnever on colab t4

import torch
from app.sana_pipeline import SanaPipeline
from torchvision.utils import save_image

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
generator = torch.Generator(device=device).manual_seed(42)

sana = SanaPipeline("/content/Sana/configs/sana_config/512ms/Sana_600M_img512.yaml")
sana.from_pretrained("https://huggingface.co/Efficient-Large-Model/Sana_1600M_512px/blob/main/checkpoints/Sana_1600M_512px.pth")
prompt = 'a cyberpunk cat with a neon sign that says "Sana"'

image = sana(
prompt=prompt,
height=128,
width=128,
guidance_scale=1.0,
pag_guidance_scale=1.0,
num_inference_steps=5,
generator=generator,
)
save_image(image, '/content/sana.png', nrow=1, normalize=True, value_range=(-1, 1))

2024-11-23 07:01:20 - [Sana] - INFO - Sampler flow_dpm-solver, flow_shift: 1.0
2024-11-23 07:01:20 - [Sana] - INFO - Inference with torch.float16, PAG guidance layer: [14]
[DC-AE] Loading model from mit-han-lab/dc-ae-f32c32-sana-1.0
Loading checkpoint shards: 100%
 2/2 [00:01<00:00,  1.87it/s]
2024-11-23 07:02:03 - [Sana] - WARNING - use pe: False, position embed interpolation: 1.0, base size: 16
2024-11-23 07:02:03 - [Sana] - WARNING - attention type: linear; ffn type: glumbconv; autocast linear attn: False

Your session crashed after using all available RAM