Hugging Face's logo

Spaces:
VAST-AI
/
MV-Adapter-T2MV-SDXL
Running on Zero

App Files Community
1
MV-Adapter-T2MV-SDXL
File size: 4,730 Bytes 
6ef620e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
import random

import gradio as gr
import numpy as np

import spaces
import torch

from inference_t2mv_sdxl import prepare_pipeline, run_pipeline


# Base model
base_model = "stabilityai/stable-diffusion-xl-base-1.0"

# Device and dtype
dtype = torch.bfloat16
device = "cuda" if torch.cuda.is_available() else "cpu"

# Hyperparameters
NUM_VIEWS = 6
HEIGHT = 768
WIDTH = 768
MAX_SEED = np.iinfo(np.int32).max

pipe = prepare_pipeline(
    base_model=base_model,
    vae_model="madebyollin/sdxl-vae-fp16-fix",
    unet_model=None,
    lora_model=None,
    adapter_path="huanngzh/mv-adapter",
    scheduler=None,
    num_views=NUM_VIEWS,
    device=device,
    dtype=dtype,
)


@spaces.GPU()
def infer(
    prompt,
    seed=42,
    randomize_seed=False,
    guidance_scale=7.0,
    num_inference_steps=50,
    negative_prompt="watermark, ugly, deformed, noisy, blurry, low contrast",
    progress=gr.Progress(track_tqdm=True),
):
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    images = run_pipeline(
        pipe,
        num_views=NUM_VIEWS,
        text=prompt,
        height=HEIGHT,
        width=WIDTH,
        num_inference_steps=num_inference_steps,
        guidance_scale=guidance_scale,
        seed=seed,
        negative_prompt=negative_prompt,
        device=device,
    )
    return images, seed


examples = {
    "stabilityai/stable-diffusion-xl-base-1.0": [
        ["An astronaut riding a horse", 42],
        ["A DSLR photo of a frog wearing a sweater", 42],
    ],
    "cagliostrolab/animagine-xl-3.1": [
        [
            "1girl, izayoi sakuya, touhou, solo, maid headdress, maid, apron, short sleeves, dress, closed mouth, white apron, serious face, upper body, masterpiece, best quality, very aesthetic, absurdres",
            0,
        ],
        [
            "1boy, male focus, ikari shinji, neon genesis evangelion, solo, serious face,(masterpiece), (best quality), (ultra-detailed), very aesthetic, illustration, disheveled hair, moist skin, intricate details",
            0,
        ],
        [
            "1girl, pink hair, pink shirts, smile, shy, masterpiece, anime",
            0,
        ],
    ],
}

css = """
#col-container {
    margin: 0 auto;
    max-width: 600px;
}
"""

with gr.Blocks(css=css) as demo:

    with gr.Column(elem_id="col-container"):
        gr.Markdown(
            f"""# MV-Adapter [Text-to-Multi-View]
Generate 768x768 multi-view images using {base_model} <br>
[[page](https://huanngzh.github.io/MV-Adapter-Page/)] [[repo](https://github.com/huanngzh/MV-Adapter)]
        """
        )

        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",
            show_label=False,
            columns=[3],
            rows=[2],
            object_fit="contain",
            height="auto",
        )

        with gr.Accordion("Advanced Settings", open=False):
            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():
                num_inference_steps = gr.Slider(
                    label="Number of inference steps",
                    minimum=1,
                    maximum=50,
                    step=1,
                    value=50,
                )

            with gr.Row():
                guidance_scale = gr.Slider(
                    label="CFG scale",
                    minimum=0.0,
                    maximum=10.0,
                    step=0.1,
                    value=7.0,
                )

            with gr.Row():
                negative_prompt = gr.Textbox(
                    label="Negative prompt",
                    placeholder="Enter your negative prompt",
                    value="watermark, ugly, deformed, noisy, blurry, low contrast",
                )

        gr.Examples(
            examples=examples[base_model],
            fn=infer,
            inputs=[prompt, seed],
            outputs=[result, seed],
            cache_examples=True,
        )

    gr.on(
        triggers=[run_button.click, prompt.submit],
        fn=infer,
        inputs=[
            prompt,
            seed,
            randomize_seed,
            guidance_scale,
            num_inference_steps,
            negative_prompt,
        ],
        outputs=[result, seed],
    )

demo.launch()