Spaces:
Running
on
Zero
Running
on
Zero
Update app_t2v.py
Browse files- app_t2v.py +62 -52
app_t2v.py
CHANGED
@@ -1,55 +1,58 @@
|
|
|
|
1 |
import os
|
2 |
os.system('pip install --upgrade --pre --extra-index-url https://download.pytorch.org/whl/nightly/cu126 "torch<2.9" spaces')
|
3 |
|
|
|
|
|
4 |
import torch
|
|
|
|
|
5 |
import gradio as gr
|
6 |
import tempfile
|
7 |
import random
|
8 |
import numpy as np
|
9 |
-
import spaces
|
10 |
-
from diffusers import WanPipeline, AutoencoderKLWan
|
11 |
-
from diffusers.utils import export_to_video
|
12 |
|
13 |
# Constants
|
14 |
MODEL_ID = "Wan-AI/Wan2.2-T2V-A14B-Diffusers"
|
15 |
-
MAX_SEED = np.iinfo(np.int32).max
|
16 |
FIXED_FPS = 16
|
17 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
"色调艳丽,过曝,静态,细节模糊不清,字幕,风格,作品,画作,画面,静止,整体发灰,"
|
19 |
"最差质量,低质量,JPEG压缩残留,丑陋的,残缺的,多余的手指,画得不好的手部,"
|
20 |
"画得不好的脸部,畸形的,毁容的,形态畸形的肢体,手指融合,静止不动的画面,杂乱的背景,三条腿,背景人很多,倒着走"
|
21 |
)
|
22 |
|
23 |
-
#
|
24 |
-
|
25 |
-
|
26 |
|
27 |
-
|
28 |
-
vae = AutoencoderKLWan.from_pretrained(
|
29 |
-
MODEL_ID, subfolder="vae", torch_dtype=torch.float32
|
30 |
-
).to(device)
|
31 |
|
32 |
-
pipe = WanPipeline.from_pretrained(
|
33 |
-
MODEL_ID, vae=vae, torch_dtype=dtype
|
34 |
-
).to(device)
|
35 |
|
36 |
-
#
|
37 |
_ = pipe(
|
38 |
prompt="warmup",
|
39 |
-
negative_prompt=
|
40 |
height=512,
|
41 |
width=768,
|
42 |
num_frames=8,
|
43 |
num_inference_steps=2,
|
44 |
-
generator=torch.Generator(device=device).manual_seed(0)
|
45 |
).frames[0]
|
46 |
|
47 |
-
#
|
48 |
-
def get_duration(prompt, negative_prompt, height, width, num_frames, guidance_scale, guidance_scale_2,
|
49 |
-
return int(
|
50 |
|
51 |
@spaces.GPU(duration=get_duration)
|
52 |
-
def
|
53 |
prompt,
|
54 |
negative_prompt,
|
55 |
height,
|
@@ -57,57 +60,64 @@ def generate_video(
|
|
57 |
num_frames,
|
58 |
guidance_scale,
|
59 |
guidance_scale_2,
|
60 |
-
|
61 |
seed,
|
62 |
-
randomize_seed
|
|
|
63 |
):
|
64 |
current_seed = random.randint(0, MAX_SEED) if randomize_seed else int(seed)
|
65 |
generator = torch.Generator(device=device).manual_seed(current_seed)
|
66 |
|
67 |
-
|
68 |
prompt=prompt,
|
69 |
negative_prompt=negative_prompt,
|
70 |
-
height=height,
|
71 |
-
width=width,
|
72 |
-
num_frames=num_frames,
|
73 |
-
guidance_scale=guidance_scale,
|
74 |
-
guidance_scale_2=guidance_scale_2,
|
75 |
-
num_inference_steps=
|
76 |
generator=generator,
|
77 |
).frames[0]
|
78 |
|
79 |
with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as tmpfile:
|
80 |
-
export_to_video(
|
81 |
return tmpfile.name, current_seed
|
82 |
|
83 |
# Gradio UI
|
84 |
with gr.Blocks() as demo:
|
85 |
-
gr.Markdown("## 🎬
|
86 |
|
87 |
with gr.Row():
|
88 |
with gr.Column():
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
-
|
99 |
-
|
100 |
-
|
|
|
|
|
101 |
|
102 |
with gr.Column():
|
103 |
video_output = gr.Video(label="Generated Video", autoplay=True, interactive=False)
|
104 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
105 |
|
106 |
-
generate_button.click(
|
107 |
-
fn=generate_video,
|
108 |
-
inputs=[prompt, negative_prompt, height, width, num_frames, guidance_scale, guidance_scale_2, num_steps, seed, randomize_seed],
|
109 |
-
outputs=[video_output, final_seed_display],
|
110 |
-
)
|
111 |
|
112 |
if __name__ == "__main__":
|
113 |
-
demo.queue().launch()
|
|
|
1 |
+
# PyTorch nightly for CUDA compatibility
|
2 |
import os
|
3 |
os.system('pip install --upgrade --pre --extra-index-url https://download.pytorch.org/whl/nightly/cu126 "torch<2.9" spaces')
|
4 |
|
5 |
+
# Imports
|
6 |
+
import spaces
|
7 |
import torch
|
8 |
+
from diffusers import WanPipeline, AutoencoderKLWan
|
9 |
+
from diffusers.utils import export_to_video
|
10 |
import gradio as gr
|
11 |
import tempfile
|
12 |
import random
|
13 |
import numpy as np
|
|
|
|
|
|
|
14 |
|
15 |
# Constants
|
16 |
MODEL_ID = "Wan-AI/Wan2.2-T2V-A14B-Diffusers"
|
|
|
17 |
FIXED_FPS = 16
|
18 |
+
MAX_SEED = np.iinfo(np.int32).max
|
19 |
+
DEFAULT_HEIGHT = 720
|
20 |
+
DEFAULT_WIDTH = 1280
|
21 |
+
MAX_FRAMES = 81
|
22 |
+
|
23 |
+
# Prompts
|
24 |
+
default_prompt_t2v = "Two anthropomorphic cats in comfy boxing gear and bright gloves fight intensely on a spotlighted stage."
|
25 |
+
default_negative_prompt = (
|
26 |
"色调艳丽,过曝,静态,细节模糊不清,字幕,风格,作品,画作,画面,静止,整体发灰,"
|
27 |
"最差质量,低质量,JPEG压缩残留,丑陋的,残缺的,多余的手指,画得不好的手部,"
|
28 |
"画得不好的脸部,畸形的,毁容的,形态畸形的肢体,手指融合,静止不动的画面,杂乱的背景,三条腿,背景人很多,倒着走"
|
29 |
)
|
30 |
|
31 |
+
# Load pipeline
|
32 |
+
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
33 |
+
dtype = torch.bfloat16 if torch.cuda.is_available() else torch.float32
|
34 |
|
35 |
+
vae = AutoencoderKLWan.from_pretrained(MODEL_ID, subfolder="vae", torch_dtype=torch.float32).to(device)
|
|
|
|
|
|
|
36 |
|
37 |
+
pipe = WanPipeline.from_pretrained(MODEL_ID, vae=vae, torch_dtype=dtype).to(device)
|
|
|
|
|
38 |
|
39 |
+
# Optional: warm-up
|
40 |
_ = pipe(
|
41 |
prompt="warmup",
|
42 |
+
negative_prompt=default_negative_prompt,
|
43 |
height=512,
|
44 |
width=768,
|
45 |
num_frames=8,
|
46 |
num_inference_steps=2,
|
47 |
+
generator=torch.Generator(device=device).manual_seed(0)
|
48 |
).frames[0]
|
49 |
|
50 |
+
# Space-aware duration helper
|
51 |
+
def get_duration(prompt, negative_prompt, height, width, num_frames, guidance_scale, guidance_scale_2, steps, seed, randomize_seed, progress):
|
52 |
+
return int(steps * 15)
|
53 |
|
54 |
@spaces.GPU(duration=get_duration)
|
55 |
+
def generate_t2v(
|
56 |
prompt,
|
57 |
negative_prompt,
|
58 |
height,
|
|
|
60 |
num_frames,
|
61 |
guidance_scale,
|
62 |
guidance_scale_2,
|
63 |
+
steps,
|
64 |
seed,
|
65 |
+
randomize_seed,
|
66 |
+
progress=gr.Progress(track_tqdm=True),
|
67 |
):
|
68 |
current_seed = random.randint(0, MAX_SEED) if randomize_seed else int(seed)
|
69 |
generator = torch.Generator(device=device).manual_seed(current_seed)
|
70 |
|
71 |
+
output_frames = pipe(
|
72 |
prompt=prompt,
|
73 |
negative_prompt=negative_prompt,
|
74 |
+
height=int(height),
|
75 |
+
width=int(width),
|
76 |
+
num_frames=int(num_frames),
|
77 |
+
guidance_scale=float(guidance_scale),
|
78 |
+
guidance_scale_2=float(guidance_scale_2),
|
79 |
+
num_inference_steps=int(steps),
|
80 |
generator=generator,
|
81 |
).frames[0]
|
82 |
|
83 |
with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as tmpfile:
|
84 |
+
export_to_video(output_frames, tmpfile.name, fps=FIXED_FPS)
|
85 |
return tmpfile.name, current_seed
|
86 |
|
87 |
# Gradio UI
|
88 |
with gr.Blocks() as demo:
|
89 |
+
gr.Markdown("## 🎬 Wan 2.2 T2V: Text-to-Video via Wan-AI")
|
90 |
|
91 |
with gr.Row():
|
92 |
with gr.Column():
|
93 |
+
prompt_input = gr.Textbox(label="Prompt", value=default_prompt_t2v)
|
94 |
+
negative_prompt_input = gr.Textbox(label="Negative Prompt", value=default_negative_prompt, lines=3)
|
95 |
+
height_slider = gr.Slider(360, 1024, step=16, value=DEFAULT_HEIGHT, label="Height")
|
96 |
+
width_slider = gr.Slider(360, 1920, step=16, value=DEFAULT_WIDTH, label="Width")
|
97 |
+
frames_slider = gr.Slider(8, MAX_FRAMES, value=MAX_FRAMES, step=1, label="Frames")
|
98 |
+
|
99 |
+
with gr.Accordion("Advanced Settings", open=False):
|
100 |
+
guidance_slider = gr.Slider(0.0, 20.0, step=0.5, value=4.0, label="Guidance Scale")
|
101 |
+
guidance2_slider = gr.Slider(0.0, 20.0, step=0.5, value=3.0, label="Guidance Scale 2")
|
102 |
+
steps_slider = gr.Slider(1, 60, step=1, value=40, label="Inference Steps")
|
103 |
+
seed_slider = gr.Slider(0, MAX_SEED, step=1, value=42, label="Seed", interactive=True)
|
104 |
+
randomize_seed_check = gr.Checkbox(label="Randomize Seed", value=True)
|
105 |
+
|
106 |
+
generate_button = gr.Button("🎥 Generate Video", variant="primary")
|
107 |
|
108 |
with gr.Column():
|
109 |
video_output = gr.Video(label="Generated Video", autoplay=True, interactive=False)
|
110 |
+
used_seed = gr.Number(label="Used Seed", interactive=False)
|
111 |
+
|
112 |
+
inputs = [
|
113 |
+
prompt_input, negative_prompt_input,
|
114 |
+
height_slider, width_slider,
|
115 |
+
frames_slider,
|
116 |
+
guidance_slider, guidance2_slider,
|
117 |
+
steps_slider, seed_slider, randomize_seed_check
|
118 |
+
]
|
119 |
|
120 |
+
generate_button.click(fn=generate_t2v, inputs=inputs, outputs=[video_output, used_seed])
|
|
|
|
|
|
|
|
|
121 |
|
122 |
if __name__ == "__main__":
|
123 |
+
demo.queue().launch(mcp_server=True)
|