File size: 6,146 Bytes
b8975f7
 
 
 
 
e984a2c
 
b8975f7
b9a20a2
b8975f7
 
e984a2c
b8975f7
 
e984a2c
3ce81f3
88a1b07
b8975f7
 
 
 
 
 
 
 
 
 
 
f87fd5b
b8975f7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3ce81f3
b8975f7
 
 
7ab092f
b8975f7
 
 
 
 
7ab092f
 
b8975f7
7ab092f
b8975f7
 
 
 
 
 
88a1b07
b8975f7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3ce81f3
 
b8975f7
ac71998
 
b8975f7
c53d852
 
 
3ce81f3
 
039e573
 
 
cb7c885
 
 
039e573
b8975f7
cb7c885
 
b8975f7
3ce81f3
 
 
 
 
 
 
 
 
ac71998
b8975f7
 
 
3ce81f3
 
cb7c885
 
3ce81f3
b8975f7
 
 
 
 
 
 
 
 
 
 
 
 
 
7ab092f
b8975f7
7ab092f
 
b8975f7
 
 
7ab092f
b8975f7
7ab092f
 
3ce81f3
b8975f7
 
 
 
 
 
 
 
 
 
 
 
3ce81f3
7ab092f
3ce81f3
 
b8975f7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3ce81f3
 
b8975f7
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
import os
import random
import uuid
import json

import gradio as gr
import numpy as np
from PIL import Image
import spaces
import torch
from diffusers import DiffusionPipeline

if not torch.cuda.is_available():
    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo may not work on CPU.</p>"

MAX_SEED = np.iinfo(np.int32).max
CACHE_EXAMPLES = torch.cuda.is_available() and os.getenv("CACHE_EXAMPLES", "1") == "1"
MAX_IMAGE_SIZE = int(os.getenv("MAX_IMAGE_SIZE", "4096"))
USE_TORCH_COMPILE = os.getenv("USE_TORCH_COMPILE", "0") == "1"
ENABLE_CPU_OFFLOAD = os.getenv("ENABLE_CPU_OFFLOAD", "0") == "1"

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

if torch.cuda.is_available():
    pipe = DiffusionPipeline.from_pretrained(
        "sd-community/sdxl-flash",
        torch_dtype=torch.float16,
        use_safetensors=True,
        add_watermarker=False
    )
    if ENABLE_CPU_OFFLOAD:
        pipe.enable_model_cpu_offload()
    else:
        pipe.to(device)       
        print("Loaded on Device!")
    
    if USE_TORCH_COMPILE:
        pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
        print("Model Compiled!")


def save_image(img):
    unique_name = str(uuid.uuid4()) + ".png"
    img.save(unique_name)
    return unique_name

def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    return seed

@spaces.GPU(duration=30, queue=False)
def generate(
    prompt: str,
    negative_prompt: str = "",
    use_negative_prompt: bool = False,
    seed: int = 0,
    width: int = 2048,
    height: int = 2048,
    guidance_scale: float = 3,
    num_inference_steps: int = 10,
    randomize_seed: bool = False,
    use_resolution_binning: bool = True,
    progress=gr.Progress(track_tqdm=True),
):
    pipe.to(device)
    seed = int(randomize_seed_fn(seed, randomize_seed))
    generator = torch.Generator().manual_seed(seed)   

    options = {
        "prompt":prompt,
        "negative_prompt":negative_prompt,
        "width":width,
        "height":height,
        "guidance_scale":guidance_scale,
        "num_inference_steps":num_inference_steps,
        "generator":generator,
        "use_resolution_binning":use_resolution_binning,
        "output_type":"pil",

    }
    
    images = pipe(**options).images

    image_paths = [save_image(img) for img in images]
    return image_paths, seed


examples = [
    "a cat eating a piece of cheese",
    "a ROBOT riding a BLUE horse on Mars, photorealistic",
    "a cartoon of a IRONMAN fighting with HULK, wall painting",
    "a cute robot artist painting on an easel, concept art",
    "Astronaut in a jungle, cold color palette, oil pastel, detailed, 8k",
    "An alien grasping a sign board contain word 'Flash', sketch, detailed",
    "Kids going to school, Anime style"
]

css = '''
.gradio-container{max-width: 560px !important}
h1{text-align:center}
footer {
    visibility: hidden
}
'''
with gr.Blocks(css=css) as demo:
    gr.Markdown("""# SDXL Flash
        ### First Image processing takes time then images generate faster.""")
    with gr.Group():
        with gr.Row():
            prompt = gr.Text(
                label="Prompt",
                show_label=False,
                max_lines=1,
                placeholder="Enter your prompt",
                container=False,
            )
            run_button = gr.Button("Run", scale=0)
        result = gr.Gallery(label="Result", columns=1)
    with gr.Accordion("Advanced options", open=False):
        with gr.Row():
            use_negative_prompt = gr.Checkbox(label="Use negative prompt", value=True)
            negative_prompt = gr.Text(
                label="Negative prompt",
                max_lines=5,
                lines=4,
                placeholder="Enter a negative prompt",
                value="(deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, NSFW",
                visible=True,
            )
        seed = gr.Slider(
            label="Seed",
            minimum=0,
            maximum=MAX_SEED,
            step=1,
            value=0,
        )
        randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
        with gr.Row(visible=True):
            width = gr.Slider(
                label="Width",
                minimum=512,
                maximum=MAX_IMAGE_SIZE,
                step=64,
                value=1536,
            )
            height = gr.Slider(
                label="Height",
                minimum=512,
                maximum=MAX_IMAGE_SIZE,
                step=64,
                value=1536,
            )
        with gr.Row():
            guidance_scale = gr.Slider(
                label="Guidance Scale",
                minimum=0.1,
                maximum=6,
                step=0.1,
                value=3.0,
            )
            num_inference_steps = gr.Slider(
                label="Number of inference steps",
                minimum=1,
                maximum=15,
                step=1,
                value=8,
            )

    gr.Examples(
        examples=examples,
        inputs=prompt,
        outputs=[result, seed],
        fn=generate,
        cache_examples=CACHE_EXAMPLES,
    )

    use_negative_prompt.change(
        fn=lambda x: gr.update(visible=x),
        inputs=use_negative_prompt,
        outputs=negative_prompt,
        api_name=False,
    )

    gr.on(
        triggers=[
            prompt.submit,
            negative_prompt.submit,
            run_button.click,
        ],
        fn=generate,
        inputs=[
            prompt,
            negative_prompt,
            use_negative_prompt,
            seed,
            width,
            height,
            guidance_scale,
            num_inference_steps,
            randomize_seed,
        ],
        outputs=[result, seed],
        api_name="run",
    )

if __name__ == "__main__":
    demo.queue(max_size=20).launch()