V0.1 by <M0n-ius>

run.py

@@ -33,7 +33,16 @@ with gr.Blocks() as Core:
                     frame_skip = gr.Slider(2, 100, 2, step=1, label="Frame Skip")
                     group_size = gr.Slider(1, 100, 10, step=1, label="Group Size")
             prompt = gr.Textbox(SYS_PROMPT, label="Prompt", lines=10, max_lines=100, show_copy_button=True)
-            result = gr.Textbox(label="Result", lines=15, max_lines=100, show_copy_button=True, interactive=False)
+            with gr.Tabs():
+                with gr.Tab("Azure"):
+                    result = gr.Textbox(label="Result", lines=15, max_lines=100, show_copy_button=True, interactive=False)
+                with gr.Tab("Google"):
+                    result_gg = gr.Textbox(label="Result", lines=15, max_lines=100, show_copy_button=True, interactive=False)
+                with gr.Tab("Anthropic"):
+                    result_ac = gr.Textbox(label="Result", lines=15, max_lines=100, show_copy_button=True, interactive=False)
+                with gr.Tab("OpenAI"):
+                    result_oai = gr.Textbox(label="Result", lines=15, max_lines=100, show_copy_button=True, interactive=False)
+
         with gr.Column(scale=2):
             with gr.Column():
                 with gr.Accordion("Model Provider", open=True):
@@ -42,16 +51,35 @@ with gr.Blocks() as Core:
                             model = gr.Dropdown(label="Model", value="GPT-4o", choices=["GPT-4o", "GPT-4v"], interactive=False)
                             key = gr.Textbox(label="Azure API Key")
                             endpoint = gr.Textbox(label="Azure Endpoint")
+                        with gr.Tab("Google"):
+                            model_gg = gr.Dropdown(label="Model", value="Gemini-1.5-Flash", choices=["Gemini-1.5-Flash", "Gemini-1.5-Pro"], interactive=False)
+                            key_gg = gr.Textbox(label="Google API Key")
+                            endpoint_gg = gr.Textbox(label="Azure Endpoint")
+                        with gr.Tab("Anthropic"):
+                            model_ac = gr.Dropdown(label="Model", value="Claude-3-Opus", choices=["Claude-3-Opus", "Claude-3-Sonnet"], interactive=False)
+                            key_ac = gr.Textbox(label="Anthropic API Key")
+                            endpoint_ac = gr.Textbox(label="Anthropic Endpoint")
+                        with gr.Tab("OpenAI"):
+                            model_oai = gr.Dropdown(label="Model", value="GPT-4o", choices=["GPT-4o", "GPT-4v"], interactive=False)
+                            key_oai = gr.Textbox(label="OpenAI API Key")
+                            endpoint_oai = gr.Textbox(label="OpenAI Endpoint")
                 with gr.Accordion("Data Source", open=True):
                     with gr.Tabs():
                         with gr.Tab("Upload"):
-                            video = gr.Video(sources="upload", show_label=False, show_share_button=False, mirror_webcam=False)
+                            video_src = gr.Video(sources="upload", show_label=False, show_share_button=False, mirror_webcam=False)
+                        with gr.Tab("HF"):
+                            video_hf = gr.Text()
+                        with gr.Tab("Onedrive"):
+                            video_od = gr.Text()
+                        with gr.Tab("Google Drive"):
+                            video_gd = gr.Text()
                 caption_button = gr.Button("Caption", variant="primary", size="lg")
         caption_button.click(
             process_caption, 
-            inputs=[prompt, temp, top_p, max_tokens, model, key, endpoint, video, frame_format, frame_skip, group_size], 
+            inputs=[prompt, temp, top_p, max_tokens, model, key, endpoint, video_src, frame_format, frame_skip, group_size], 
             outputs=[result, info, frame]
         )
 
 if __name__ == "__main__":
     Core.launch()
+