init

app.py

@@ -262,7 +262,7 @@ def generate_image(prompt_scene,
         images = rearrange(images, 'b c h w -> b h w c')
         images = torch.clamp(127.5 * images + 128.0, 0, 255).to("cpu", dtype=torch.uint8).numpy()
         ret_images = [Image.fromarray(image) for image in images]
-        return ret_images, output_reasoning
+        return ret_images, output_reasoning[0]
 
 
 # Gradio interface
@@ -382,7 +382,7 @@ with gr.Blocks(css=custom_css) as demo:
     
         image_output = gr.Gallery(label="Generated images", columns=4, rows=1)
         
-        output_reasoning = gr.Textbox(label="Response")
+        output_reasoning = gr.Textbox(label="Thinking Process")
     
         examples_t2i = gr.Examples(
             label="Prompt examples",
@@ -401,10 +401,21 @@ with gr.Blocks(css=custom_css) as demo:
         gr.Markdown(value="## Camera Understanding")
         image_input = gr.Image()
         
-        with gr.Accordion("Settings", open=True):
-            with gr.Row():
-                thinking_und = gr.Checkbox(label="Thinking", value=False)
-                und_seed_input = gr.Number(label="Seed (Optional)", precision=0, value=42)
+        with gr.Accordion("Settings", open=True):            
+            with gr.Row(equal_height=True):
+                thinking_und = gr.Radio(
+                    ["Thinking"],
+                    label=None,
+                    value=None,
+                    show_label=False,
+                    interactive=True
+                )
+
+                seed_input = gr.Number(
+                    label="Seed (Optional)",
+                    precision=0,
+                    value=42
+                )
 
         understanding_button = gr.Button("Chat")
         understanding_output = gr.Textbox(label="Response")