patrickvonplaten commited on
Commit
8c4798e
1 Parent(s): 5fe03ce

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +24 -6
README.md CHANGED
@@ -46,16 +46,34 @@ The exact dependencies is got using `pip freeze` and can be found in `exact_requ
46
 
47
  Check our jupyter notebooks with examples in `./examples` folder
48
 
49
- ### 1. text2image
50
 
51
  ```python
52
- from diffusers import KandinskyV3Pipeline
53
- import torch
54
 
55
- pipe = KandinskyV3Img2ImgPipeline.from_pretrained('kandinsky-community/kandinsky-3', torch_dtype=torch.float16)
56
- pipe = pipe.to('cuda')
 
 
57
 
58
- image = pipe("A cute corgi lives in a house made out of sushi.")
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59
  ```
60
 
61
  ## Examples of generations
 
46
 
47
  Check our jupyter notebooks with examples in `./examples` folder
48
 
49
+ ### Text-2-Image
50
 
51
  ```python
52
+ from diffusers import AutoPipelineForText2Image
 
53
 
54
+ pipe = AutoPipelineForText2Image.from_pretrained("kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16)
55
+ pipe.enable_model_cpu_offload()
56
+
57
+ prompt = "A photograph of the inside of a subway train. There are raccoons sitting on the seats. One of them is reading a newspaper. The window shows the city in the background."
58
 
59
+ generator = torch.Generator(device="cpu").manual_seed(0)
60
+ image = pipe(prompt, num_inference_steps=25, generator=generator).images[0]
61
+ ```
62
+
63
+ ### Image-2-Image
64
+
65
+ ```python
66
+ from diffusers import AutoPipelineForImage2Image
67
+ from diffusers.utils import load_image
68
+
69
+ pipe = AutoPipelineForImage2Image.from_pretrained("kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16)
70
+ pipe.enable_model_cpu_offload()
71
+
72
+ prompt = "A painting of the inside of a subway train with tiny raccoons."
73
+ image = load_image("https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky3/t2i.png")
74
+
75
+ generator = torch.Generator(device="cpu").manual_seed(0)
76
+ image = pipe(prompt, image=image, strength=0.75, num_inference_steps=25, generator=generator).images[0]
77
  ```
78
 
79
  ## Examples of generations