Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,24 +1,19 @@
|
|
| 1 |
import gradio as gr
|
| 2 |
from gradio_client import Client, handle_file
|
| 3 |
-
from google import genai
|
| 4 |
-
from google.genai import types
|
| 5 |
import os
|
| 6 |
from typing import Optional, List
|
| 7 |
from huggingface_hub import whoami
|
| 8 |
from PIL import Image
|
| 9 |
from io import BytesIO
|
| 10 |
import tempfile
|
| 11 |
-
import
|
| 12 |
|
| 13 |
# --- Google Gemini API Configuration ---
|
| 14 |
GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY", "")
|
| 15 |
if not GOOGLE_API_KEY:
|
| 16 |
raise ValueError("GOOGLE_API_KEY environment variable not set.")
|
| 17 |
-
|
| 18 |
-
client = genai.Client(
|
| 19 |
-
api_key=os.environ.get("GOOGLE_API_KEY"),
|
| 20 |
-
)
|
| 21 |
-
|
| 22 |
GEMINI_MODEL_NAME = 'gemini-2.5-flash-image-preview'
|
| 23 |
|
| 24 |
def verify_pro_status(token: Optional[gr.OAuthToken]) -> bool:
|
|
@@ -27,12 +22,7 @@ def verify_pro_status(token: Optional[gr.OAuthToken]) -> bool:
|
|
| 27 |
return False
|
| 28 |
try:
|
| 29 |
user_info = whoami(token=token.token)
|
| 30 |
-
|
| 31 |
-
return True
|
| 32 |
-
orgs = user_info.get("orgs", [])
|
| 33 |
-
if any(org.get("isEnterprise", False) for org in orgs):
|
| 34 |
-
return True
|
| 35 |
-
return False
|
| 36 |
except Exception as e:
|
| 37 |
print(f"Could not verify user's PRO/Enterprise status: {e}")
|
| 38 |
return False
|
|
@@ -40,104 +30,133 @@ def verify_pro_status(token: Optional[gr.OAuthToken]) -> bool:
|
|
| 40 |
def _extract_image_data_from_response(response) -> Optional[bytes]:
|
| 41 |
"""Helper to extract image data from the model's response."""
|
| 42 |
if hasattr(response, 'candidates') and response.candidates:
|
| 43 |
-
for
|
| 44 |
-
if hasattr(
|
| 45 |
-
|
| 46 |
-
if hasattr(part, 'inline_data') and hasattr(part.inline_data, 'data'):
|
| 47 |
-
return part.inline_data.data
|
| 48 |
return None
|
| 49 |
|
| 50 |
-
def
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
"""
|
| 56 |
-
|
| 57 |
-
Returns: (output_image_path, video_button_visible, video_output_visible)
|
| 58 |
"""
|
| 59 |
-
|
| 60 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
try:
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
contents.append(prompt)
|
| 73 |
-
|
| 74 |
-
response = None
|
| 75 |
-
for attempt in range(0, 3):
|
| 76 |
-
try:
|
| 77 |
-
response = client.models.generate_content(
|
| 78 |
-
model=GEMINI_MODEL_NAME,
|
| 79 |
-
contents=contents,
|
| 80 |
-
)
|
| 81 |
-
break
|
| 82 |
-
except Exception as e:
|
| 83 |
-
if attempt == 2:
|
| 84 |
-
raise gr.Error(f"The Gemini API returned an error: {e}")
|
| 85 |
-
time.sleep(1)
|
| 86 |
-
|
| 87 |
image_data = _extract_image_data_from_response(response)
|
|
|
|
| 88 |
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
# Save the generated image to a temporary file to return its path
|
| 93 |
-
pil_image = Image.open(BytesIO(image_data))
|
| 94 |
-
with tempfile.NamedTemporaryFile(delete=False, suffix=".png") as tmpfile:
|
| 95 |
-
pil_image.save(tmpfile.name)
|
| 96 |
-
output_path = tmpfile.name
|
| 97 |
|
| 98 |
-
|
| 99 |
-
|
| 100 |
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
|
|
|
|
|
|
|
|
|
| 104 |
except Exception as e:
|
| 105 |
raise gr.Error(f"Image generation failed: {e}")
|
| 106 |
|
| 107 |
-
def
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
output_image:
|
| 111 |
-
oauth_token: Optional[gr.OAuthToken] = None
|
| 112 |
-
) -> tuple:
|
| 113 |
-
"""
|
| 114 |
-
Creates a video transition between the input and output images.
|
| 115 |
-
Returns: (video_path, video_visible)
|
| 116 |
-
"""
|
| 117 |
-
if not verify_pro_status(oauth_token):
|
| 118 |
-
raise gr.Error("Access Denied. This service is for PRO users only.")
|
| 119 |
-
|
| 120 |
-
if not input_image_gallery or not output_image:
|
| 121 |
-
raise gr.Error("Both input and output images are required for video creation.")
|
| 122 |
-
|
| 123 |
try:
|
| 124 |
-
|
| 125 |
-
|
| 126 |
-
input_image_path = input_image_gallery[0][0]
|
| 127 |
-
|
| 128 |
-
result = video_client.predict(
|
| 129 |
-
start_image_pil=handle_file(input_image_path),
|
| 130 |
-
end_image_pil=handle_file(output_image),
|
| 131 |
-
prompt=prompt_input,
|
| 132 |
-
api_name="/generate_video"
|
| 133 |
-
)
|
| 134 |
-
print(result)
|
| 135 |
-
return result[0]["video"]
|
| 136 |
-
|
| 137 |
except Exception as e:
|
| 138 |
raise gr.Error(f"Video creation failed: {e}")
|
| 139 |
|
| 140 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 141 |
css = '''
|
| 142 |
#sub_title{margin-top: -35px !important}
|
| 143 |
.tab-wrapper{margin-bottom: -33px !important}
|
|
@@ -158,77 +177,68 @@ with gr.Blocks(theme=gr.themes.Citrus(), css=css) as demo:
|
|
| 158 |
<img class="logo-dark" src='https://huggingface.co/spaces/multimodalart/nano-banana/resolve/main/nano_banana_pros.png' style='margin: 0 auto; max-width: 500px' />
|
| 159 |
<img class="logo-light" src='https://huggingface.co/spaces/multimodalart/nano-banana/resolve/main/nano_banana_pros_light.png' style='margin: 0 auto; max-width: 500px' />
|
| 160 |
''')
|
| 161 |
-
|
| 162 |
gr.HTML("<h3 style='text-align:center'>Hugging Face PRO users can use Google's Nano Banana (Gemini 2.5 Flash Image Preview) on this Space. <a href='http://huggingface.co/subscribe/pro?source=nana_banana' target='_blank'>Subscribe to PRO</a></h3>", elem_id="sub_title")
|
| 163 |
-
|
| 164 |
pro_message = gr.Markdown(visible=False)
|
| 165 |
main_interface = gr.Column(visible=False)
|
|
|
|
| 166 |
|
| 167 |
with main_interface:
|
| 168 |
with gr.Row():
|
| 169 |
with gr.Column(scale=1):
|
| 170 |
-
|
| 171 |
-
|
| 172 |
-
|
| 173 |
-
file_types=["image"],
|
| 174 |
-
height="auto"
|
| 175 |
-
)
|
| 176 |
-
|
| 177 |
-
prompt_input = gr.Textbox(
|
| 178 |
-
label="Prompt",
|
| 179 |
-
placeholder="Turns this photo into a masterpiece"
|
| 180 |
-
)
|
| 181 |
-
generate_button = gr.Button("Generate", variant="primary")
|
| 182 |
-
|
| 183 |
with gr.Column(scale=1):
|
| 184 |
output_image = gr.Image(label="Output", interactive=False, elem_id="output", type="filepath")
|
| 185 |
-
use_image_button = gr.Button("♻️ Use this Image for Next Edit")
|
| 186 |
-
|
|
|
|
|
|
|
| 187 |
with gr.Group(visible=False) as video_group:
|
| 188 |
video_output = gr.Video(label="Generated Video", show_download_button=True, autoplay=True)
|
| 189 |
gr.Markdown("Generate more with [Wan 2.2 first-last-frame](https://huggingface.co/spaces/multimodalart/wan-2-2-first-last-frame)", elem_id="wan_ad")
|
| 190 |
gr.Markdown("## Thank you for being a PRO! 🤗")
|
| 191 |
-
|
| 192 |
login_button = gr.LoginButton()
|
| 193 |
-
|
| 194 |
-
# --- Event Handlers ---
|
| 195 |
gr.on(
|
| 196 |
triggers=[generate_button.click, prompt_input.submit],
|
| 197 |
-
fn=lambda: [gr.update(visible=False), gr.update(visible=False)],
|
| 198 |
-
inputs=[],
|
| 199 |
-
outputs=[create_video_button, video_group],
|
| 200 |
-
).then(
|
| 201 |
fn=unified_image_generator,
|
| 202 |
-
inputs=[prompt_input, image_input_gallery],
|
| 203 |
-
outputs=[output_image, create_video_button, video_group]
|
| 204 |
)
|
| 205 |
|
| 206 |
use_image_button.click(
|
| 207 |
-
lambda
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 208 |
inputs=[output_image],
|
| 209 |
-
outputs=[image_input_gallery]
|
| 210 |
)
|
| 211 |
-
|
| 212 |
-
# Video creation handler
|
| 213 |
create_video_button.click(
|
| 214 |
-
fn=lambda: gr.update(visible=True),
|
| 215 |
-
inputs=[],
|
| 216 |
-
outputs=[video_group],
|
| 217 |
).then(
|
| 218 |
-
fn=
|
| 219 |
inputs=[image_input_gallery, prompt_input, output_image],
|
| 220 |
-
outputs=[video_output],
|
| 221 |
)
|
| 222 |
|
| 223 |
-
|
| 224 |
-
|
| 225 |
-
|
| 226 |
-
|
| 227 |
-
|
| 228 |
-
|
| 229 |
-
|
| 230 |
-
|
| 231 |
-
|
|
|
|
|
|
|
| 232 |
else:
|
| 233 |
message = (
|
| 234 |
"## ✨ Exclusive Access for PRO Users\n\n"
|
|
@@ -237,9 +247,7 @@ with gr.Blocks(theme=gr.themes.Citrus(), css=css) as demo:
|
|
| 237 |
"### [**Become a PRO Today!**](http://huggingface.co/subscribe/pro?source=nana_banana)"
|
| 238 |
)
|
| 239 |
return gr.update(visible=False), gr.update(visible=True, value=message)
|
| 240 |
-
|
| 241 |
demo.load(control_access, inputs=None, outputs=[main_interface, pro_message])
|
| 242 |
|
| 243 |
if __name__ == "__main__":
|
| 244 |
-
demo.queue(max_size=None, default_concurrency_limit=None)
|
| 245 |
-
demo.launch()
|
|
|
|
| 1 |
import gradio as gr
|
| 2 |
from gradio_client import Client, handle_file
|
| 3 |
+
from google import genai
|
|
|
|
| 4 |
import os
|
| 5 |
from typing import Optional, List
|
| 6 |
from huggingface_hub import whoami
|
| 7 |
from PIL import Image
|
| 8 |
from io import BytesIO
|
| 9 |
import tempfile
|
| 10 |
+
import ffmpeg
|
| 11 |
|
| 12 |
# --- Google Gemini API Configuration ---
|
| 13 |
GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY", "")
|
| 14 |
if not GOOGLE_API_KEY:
|
| 15 |
raise ValueError("GOOGLE_API_KEY environment variable not set.")
|
| 16 |
+
client = genai.Client(api_key=os.environ.get("GOOGLE_API_KEY"))
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
GEMINI_MODEL_NAME = 'gemini-2.5-flash-image-preview'
|
| 18 |
|
| 19 |
def verify_pro_status(token: Optional[gr.OAuthToken]) -> bool:
|
|
|
|
| 22 |
return False
|
| 23 |
try:
|
| 24 |
user_info = whoami(token=token.token)
|
| 25 |
+
return user_info.get("isPro", False) or any(org.get("isEnterprise", False) for org in user_info.get("orgs", []))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
except Exception as e:
|
| 27 |
print(f"Could not verify user's PRO/Enterprise status: {e}")
|
| 28 |
return False
|
|
|
|
| 30 |
def _extract_image_data_from_response(response) -> Optional[bytes]:
|
| 31 |
"""Helper to extract image data from the model's response."""
|
| 32 |
if hasattr(response, 'candidates') and response.candidates:
|
| 33 |
+
for part in response.candidates[0].content.parts:
|
| 34 |
+
if hasattr(part, 'inline_data') and hasattr(part.inline_data, 'data'):
|
| 35 |
+
return part.inline_data.data
|
|
|
|
|
|
|
| 36 |
return None
|
| 37 |
|
| 38 |
+
def _get_framerate(video_path: str) -> float:
|
| 39 |
+
"""Instantly gets the framerate of a video using ffprobe."""
|
| 40 |
+
probe = ffmpeg.probe(video_path)
|
| 41 |
+
video_stream = next((stream for stream in probe['streams'] if stream['codec_type'] == 'video'), None)
|
| 42 |
+
if video_stream is None:
|
| 43 |
+
raise ValueError("Could not find video stream in the file.")
|
| 44 |
+
return eval(video_stream['avg_frame_rate'])
|
| 45 |
+
|
| 46 |
+
def _trim_first_frame_fast(video_path: str) -> str:
|
| 47 |
+
"""
|
| 48 |
+
Removes exactly the first frame of a video without re-encoding.
|
| 49 |
+
This is the frame-accurate and fast method.
|
| 50 |
+
"""
|
| 51 |
+
gr.Info("Preparing video segment...")
|
| 52 |
+
with tempfile.NamedTemporaryFile(delete=False, suffix=".mp4") as tmp_output_file:
|
| 53 |
+
output_path = tmp_output_file.name
|
| 54 |
+
|
| 55 |
+
try:
|
| 56 |
+
framerate = _get_framerate(video_path)
|
| 57 |
+
if framerate == 0: raise ValueError("Framerate cannot be zero.")
|
| 58 |
+
start_time = 1 / framerate
|
| 59 |
+
|
| 60 |
+
# The key is placing -ss AFTER -i for accuracy, combined with -c copy for speed.
|
| 61 |
+
(
|
| 62 |
+
ffmpeg
|
| 63 |
+
.input(video_path, ss=start_time)
|
| 64 |
+
.output(output_path, c='copy', avoid_negative_ts='make_zero')
|
| 65 |
+
.run(overwrite_output=True, quiet=True)
|
| 66 |
+
)
|
| 67 |
+
return output_path
|
| 68 |
+
except Exception as e:
|
| 69 |
+
raise RuntimeError(f"FFmpeg trim error: {e}")
|
| 70 |
+
|
| 71 |
+
def _combine_videos_simple(video1_path: str, video2_path: str) -> str:
|
| 72 |
"""
|
| 73 |
+
Combines two videos using the fast concat demuxer. Assumes video2 is already trimmed.
|
|
|
|
| 74 |
"""
|
| 75 |
+
gr.Info("Stitching videos...")
|
| 76 |
+
with tempfile.NamedTemporaryFile(delete=False, mode='w', suffix=".txt") as tmp_list_file:
|
| 77 |
+
tmp_list_file.write(f"file '{os.path.abspath(video1_path)}'\n")
|
| 78 |
+
tmp_list_file.write(f"file '{os.path.abspath(video2_path)}'\n")
|
| 79 |
+
list_file_path = tmp_list_file.name
|
| 80 |
+
|
| 81 |
+
with tempfile.NamedTemporaryFile(delete=False, suffix=".mp4") as tmp_output_file:
|
| 82 |
+
output_path = tmp_output_file.name
|
| 83 |
|
| 84 |
try:
|
| 85 |
+
(
|
| 86 |
+
ffmpeg
|
| 87 |
+
.input(list_file_path, format='concat', safe=0)
|
| 88 |
+
.output(output_path, c='copy')
|
| 89 |
+
.run(overwrite_output=True, quiet=True)
|
| 90 |
+
)
|
| 91 |
+
return output_path
|
| 92 |
+
except ffmpeg.Error as e:
|
| 93 |
+
raise RuntimeError(f"FFmpeg combine error: {e.stderr.decode()}")
|
| 94 |
+
finally:
|
| 95 |
+
if os.path.exists(list_file_path):
|
| 96 |
+
os.remove(list_file_path)
|
| 97 |
+
|
| 98 |
+
def _generate_video_segment(input_image_path: str, output_image_path: str, prompt: str, token: str) -> str:
|
| 99 |
+
"""Generates a single video segment using the external service."""
|
| 100 |
+
gr.Info("Generating new video segment...")
|
| 101 |
+
video_client = Client("multimodalart/wan-2-2-first-last-frame", hf_token=token)
|
| 102 |
+
result = video_client.predict(
|
| 103 |
+
start_image_pil=handle_file(input_image_path),
|
| 104 |
+
end_image_pil=handle_file(output_image_path),
|
| 105 |
+
prompt=prompt, api_name="/generate_video"
|
| 106 |
+
)
|
| 107 |
+
return result[0]["video"]
|
| 108 |
+
|
| 109 |
+
def unified_image_generator(prompt: str, images: Optional[List[str]], previous_video_path: Optional[str], oauth_token: Optional[gr.OAuthToken]) -> tuple:
|
| 110 |
+
"""
|
| 111 |
+
Handles image generation and determines the visibility of video creation buttons.
|
| 112 |
+
"""
|
| 113 |
+
if not verify_pro_status(oauth_token): raise gr.Error("Access Denied.")
|
| 114 |
+
try:
|
| 115 |
+
contents = [Image.open(image_path[0]) for image_path in images] if images else []
|
| 116 |
contents.append(prompt)
|
| 117 |
+
response = client.models.generate_content(model=GEMINI_MODEL_NAME, contents=contents)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 118 |
image_data = _extract_image_data_from_response(response)
|
| 119 |
+
if not image_data: raise ValueError("No image data in response.")
|
| 120 |
|
| 121 |
+
with tempfile.NamedTemporaryFile(delete=False, suffix=".png") as tmp:
|
| 122 |
+
Image.open(BytesIO(image_data)).save(tmp.name)
|
| 123 |
+
output_path = tmp.name
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 124 |
|
| 125 |
+
can_create_video = bool(images and len(images) == 1)
|
| 126 |
+
can_extend_video = can_create_video and bool(previous_video_path)
|
| 127 |
|
| 128 |
+
return (
|
| 129 |
+
output_path,
|
| 130 |
+
gr.update(visible=can_create_video),
|
| 131 |
+
gr.update(visible=can_extend_video),
|
| 132 |
+
gr.update(visible=False)
|
| 133 |
+
)
|
| 134 |
except Exception as e:
|
| 135 |
raise gr.Error(f"Image generation failed: {e}")
|
| 136 |
|
| 137 |
+
def create_new_video(input_image_gallery: List[str], prompt_input: str, output_image: str, oauth_token: Optional[gr.OAuthToken]) -> tuple:
|
| 138 |
+
"""Starts a NEW video chain, overwriting any previous video state."""
|
| 139 |
+
if not verify_pro_status(oauth_token): raise gr.Error("Access Denied.")
|
| 140 |
+
if not input_image_gallery or not output_image: raise gr.Error("Input/output images required.")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 141 |
try:
|
| 142 |
+
new_segment_path = _generate_video_segment(input_image_gallery[0][0], output_image, prompt_input, oauth_token.token)
|
| 143 |
+
return new_segment_path, new_segment_path
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 144 |
except Exception as e:
|
| 145 |
raise gr.Error(f"Video creation failed: {e}")
|
| 146 |
|
| 147 |
+
def extend_existing_video(input_image_gallery: List[str], prompt_input: str, output_image: str, previous_video_path: str, oauth_token: Optional[gr.OAuthToken]) -> tuple:
|
| 148 |
+
"""Extends an existing video with a new segment."""
|
| 149 |
+
if not verify_pro_status(oauth_token): raise gr.Error("Access Denied.")
|
| 150 |
+
if not previous_video_path: raise gr.Error("No previous video to extend.")
|
| 151 |
+
if not input_image_gallery or not output_image: raise gr.Error("Input/output images required.")
|
| 152 |
+
try:
|
| 153 |
+
new_segment_path = _generate_video_segment(input_image_gallery[0][0], output_image, prompt_input, oauth_token.token)
|
| 154 |
+
trimmed_segment_path = _trim_first_frame_fast(new_segment_path)
|
| 155 |
+
final_video_path = _combine_videos_simple(previous_video_path, trimmed_segment_path)
|
| 156 |
+
return final_video_path, final_video_path
|
| 157 |
+
except Exception as e:
|
| 158 |
+
raise gr.Error(f"Video extension failed: {e}")
|
| 159 |
+
|
| 160 |
css = '''
|
| 161 |
#sub_title{margin-top: -35px !important}
|
| 162 |
.tab-wrapper{margin-bottom: -33px !important}
|
|
|
|
| 177 |
<img class="logo-dark" src='https://huggingface.co/spaces/multimodalart/nano-banana/resolve/main/nano_banana_pros.png' style='margin: 0 auto; max-width: 500px' />
|
| 178 |
<img class="logo-light" src='https://huggingface.co/spaces/multimodalart/nano-banana/resolve/main/nano_banana_pros_light.png' style='margin: 0 auto; max-width: 500px' />
|
| 179 |
''')
|
|
|
|
| 180 |
gr.HTML("<h3 style='text-align:center'>Hugging Face PRO users can use Google's Nano Banana (Gemini 2.5 Flash Image Preview) on this Space. <a href='http://huggingface.co/subscribe/pro?source=nana_banana' target='_blank'>Subscribe to PRO</a></h3>", elem_id="sub_title")
|
|
|
|
| 181 |
pro_message = gr.Markdown(visible=False)
|
| 182 |
main_interface = gr.Column(visible=False)
|
| 183 |
+
previous_video_state = gr.State(None)
|
| 184 |
|
| 185 |
with main_interface:
|
| 186 |
with gr.Row():
|
| 187 |
with gr.Column(scale=1):
|
| 188 |
+
image_input_gallery = gr.Gallery(label="Upload one or more images here. Leave empty for text-to-image", file_types=["image"], height="auto")
|
| 189 |
+
prompt_input = gr.Textbox(label="Prompt", placeholder="Turns this photo into a masterpiece")
|
| 190 |
+
generate_button = gr.Button("Generate", variant="primary")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 191 |
with gr.Column(scale=1):
|
| 192 |
output_image = gr.Image(label="Output", interactive=False, elem_id="output", type="filepath")
|
| 193 |
+
use_image_button = gr.Button("♻️ Use this Image for Next Edit", variant="primary")
|
| 194 |
+
with gr.Row():
|
| 195 |
+
create_video_button = gr.Button("Create video between the two images 🎥", variant="secondary", visible=False)
|
| 196 |
+
extend_video_button = gr.Button("Extend previous video with new scene 🎞️", variant="secondary", visible=False)
|
| 197 |
with gr.Group(visible=False) as video_group:
|
| 198 |
video_output = gr.Video(label="Generated Video", show_download_button=True, autoplay=True)
|
| 199 |
gr.Markdown("Generate more with [Wan 2.2 first-last-frame](https://huggingface.co/spaces/multimodalart/wan-2-2-first-last-frame)", elem_id="wan_ad")
|
| 200 |
gr.Markdown("## Thank you for being a PRO! 🤗")
|
| 201 |
+
|
| 202 |
login_button = gr.LoginButton()
|
| 203 |
+
|
|
|
|
| 204 |
gr.on(
|
| 205 |
triggers=[generate_button.click, prompt_input.submit],
|
|
|
|
|
|
|
|
|
|
|
|
|
| 206 |
fn=unified_image_generator,
|
| 207 |
+
inputs=[prompt_input, image_input_gallery, previous_video_state],
|
| 208 |
+
outputs=[output_image, create_video_button, extend_video_button, video_group]
|
| 209 |
)
|
| 210 |
|
| 211 |
use_image_button.click(
|
| 212 |
+
fn=lambda img: (
|
| 213 |
+
[img] if img else None,
|
| 214 |
+
None,
|
| 215 |
+
gr.update(visible=False),
|
| 216 |
+
gr.update(visible=False),
|
| 217 |
+
gr.update(visible=False)
|
| 218 |
+
),
|
| 219 |
inputs=[output_image],
|
| 220 |
+
outputs=[image_input_gallery, output_image, create_video_button, extend_video_button, video_group]
|
| 221 |
)
|
| 222 |
+
|
|
|
|
| 223 |
create_video_button.click(
|
| 224 |
+
fn=lambda: gr.update(visible=True), outputs=[video_group]
|
|
|
|
|
|
|
| 225 |
).then(
|
| 226 |
+
fn=create_new_video,
|
| 227 |
inputs=[image_input_gallery, prompt_input, output_image],
|
| 228 |
+
outputs=[video_output, previous_video_state],
|
| 229 |
)
|
| 230 |
|
| 231 |
+
extend_video_button.click(
|
| 232 |
+
fn=lambda: gr.update(visible=True), outputs=[video_group]
|
| 233 |
+
).then(
|
| 234 |
+
fn=extend_existing_video,
|
| 235 |
+
inputs=[image_input_gallery, prompt_input, output_image, previous_video_state],
|
| 236 |
+
outputs=[video_output, previous_video_state],
|
| 237 |
+
)
|
| 238 |
+
|
| 239 |
+
def control_access(profile: Optional[gr.OAuthProfile] = None, oauth_token: Optional[gr.OAuthToken] = None):
|
| 240 |
+
if not profile: return gr.update(visible=False), gr.update(visible=False)
|
| 241 |
+
if verify_pro_status(oauth_token): return gr.update(visible=True), gr.update(visible=False)
|
| 242 |
else:
|
| 243 |
message = (
|
| 244 |
"## ✨ Exclusive Access for PRO Users\n\n"
|
|
|
|
| 247 |
"### [**Become a PRO Today!**](http://huggingface.co/subscribe/pro?source=nana_banana)"
|
| 248 |
)
|
| 249 |
return gr.update(visible=False), gr.update(visible=True, value=message)
|
|
|
|
| 250 |
demo.load(control_access, inputs=None, outputs=[main_interface, pro_message])
|
| 251 |
|
| 252 |
if __name__ == "__main__":
|
| 253 |
+
demo.queue(max_size=None, default_concurrency_limit=None).launch()
|
|
|