File size: 4,659 Bytes
65fd06d
 
 
 
 
 
 
 
72c347e
65fd06d
 
 
7f11b82
65fd06d
72c347e
 
 
65fd06d
 
72c347e
7db3ca3
72c347e
 
 
 
 
65fd06d
 
7f11b82
65fd06d
 
 
 
 
 
 
 
 
 
 
 
72c347e
 
88cc598
 
0810225
 
88cc598
65fd06d
 
 
 
 
7db3ca3
 
 
 
 
 
 
 
65fd06d
108abb9
895e905
 
 
9aa1c3d
 
895e905
 
108abb9
 
 
 
 
 
 
 
 
 
 
 
 
65fd06d
108abb9
 
 
 
 
 
 
 
 
 
 
 
 
65fd06d
108abb9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
72c347e
 
108abb9
72c347e
108abb9
 
72c347e
 
108abb9
72c347e
108abb9
65fd06d
 
 
8790f79
 
 
65fd06d
 
 
 
 
 
 
 
 
7db3ca3
 
7f11b82
65fd06d
 
 
 
8790f79
 
 
65fd06d
 
 
 
 
 
8790f79
4fec0c4
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
#!/usr/bin/env python

import os
import random
import gradio as gr
import numpy as np
import PIL.Image
import torch
from diffusers import LCMScheduler, AutoPipelineForText2Image

MAX_SEED = np.iinfo(np.int32).max
MAX_IMAGE_SIZE = int(os.getenv('MAX_IMAGE_SIZE', '1024'))
SECRET_TOKEN = os.getenv('SECRET_TOKEN', 'default_secret')

MODEL_ID = "segmind/SSD-1B"
ADAPTER_ID = "latent-consistency/lcm-lora-ssd-1b"

device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
if torch.cuda.is_available():
    pipe = AutoPipelineForText2Image.from_pretrained(MODEL_ID, torch_dtype=torch.float16, variant="fp16")
    pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
    pipe.to("cuda")

    # load and fuse
    pipe.load_lora_weights(ADAPTER_ID)
    pipe.fuse_lora()
else:
    pipe = None
    
def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    return seed


def generate(prompt: str,
             negative_prompt: str = '',
             use_negative_prompt: bool = False,
             seed: int = 0,
             width: int = 1024,
             height: int = 1024,
             guidance_scale: float = 0.0,
             num_inference_steps: int = 4,
             secret_token: str = '') -> PIL.Image.Image:
    if secret_token != SECRET_TOKEN:
        raise gr.Error(
            f'Invalid secret token. Please fork the original space if you want to use it for yourself.')
        
    generator = torch.Generator().manual_seed(seed)

    if not use_negative_prompt:
        negative_prompt = None  # type: ignore

    return pipe(prompt=prompt,
                negative_prompt=negative_prompt,
                width=width,
                height=height,
                guidance_scale=guidance_scale,
                num_inference_steps=num_inference_steps,
                generator=generator,
                output_type='pil').images[0]

with gr.Blocks() as demo:
    gr.HTML("""
    <div style="z-index: 100; position: fixed; top: 0px; right: 0px; left: 0px; bottom: 0px; width: 100%; height: 100%; background: white; display: flex; align-items: center; justify-content: center; color: black;">
        <div style="text-align: center; color: black;">
            <p style="color: black;">This space is a REST API to programmatically generate images using LCM-SSD-1B.</p>
            <p style="color: black;">It is not meant to be directly used through a user interface, but using code and an access key.</p>
        </div>
    </div>""")
    secret_token = gr.Text(
        label='Secret Token',
        max_lines=1,
        placeholder='Enter your secret token',
    )
    prompt = gr.Text(
        label='Prompt',
        show_label=False,
        max_lines=1,
        placeholder='Enter your prompt',
        container=False,
    )
    result = gr.Image(label='Result', show_label=False)

    use_negative_prompt = gr.Checkbox(label='Use negative prompt', value=False)
    negative_prompt = gr.Text(
        label='Negative prompt',
        max_lines=1,
        placeholder='Enter a negative prompt',
        visible=False,
    )
    seed = gr.Slider(label='Seed',
                    minimum=0,
                    maximum=MAX_SEED,
                    step=1,
                    value=0)
    randomize_seed = gr.Checkbox(label='Randomize seed', value=True)

    width = gr.Slider(
        label='Width',
        minimum=256,
        maximum=MAX_IMAGE_SIZE,
        step=32,
        value=1024,
    )
    height = gr.Slider(
        label='Height',
        minimum=256,
        maximum=MAX_IMAGE_SIZE,
        step=32,
        value=1024,
    )
    guidance_scale = gr.Slider(
        label='Guidance scale',
        minimum=0,
        maximum=2,
        step=0.1,
        value=0.0)
    num_inference_steps = gr.Slider(
        label='Number of inference steps',
        minimum=1,
        maximum=8,
        step=1,
        value=4)
    
    use_negative_prompt.change(
        fn=lambda x: gr.update(visible=x),
        inputs=use_negative_prompt,
        outputs=negative_prompt,
        queue=False,
        api_name=False,
    )

    inputs = [
        prompt,
        negative_prompt,
        use_negative_prompt,
        seed,
        width,
        height,
        guidance_scale,
        num_inference_steps,
        secret_token,
    ]
    prompt.submit(
        fn=randomize_seed_fn,
        inputs=[seed, randomize_seed],
        outputs=seed,
        queue=False,
        api_name=False,
    ).then(
        fn=generate,
        inputs=inputs,
        outputs=result,
        api_name='run',
    )

demo.queue(max_size=16).launch()