Update app.py

app.py

@@ -2,11 +2,9 @@ import gradio as gr
 from gradio_client import Client, handle_file
 from huggingface_hub import InferenceClient
 
-# Clients for different models
 moondream_client = Client("vikhyatk/moondream2")
 qwq_client = InferenceClient("Qwen/QwQ-32B-Preview")
 
-# Function to describe the image
 def describe_image(image, user_message):
     result = moondream_client.predict(
         img=handle_file(image),
@@ -15,6 +13,7 @@ def describe_image(image, user_message):
     )
     
     description = result
+
     user_message = description + "\n" + user_message 
 
     qwq_result = qwq_client.chat_completion(
@@ -26,68 +25,26 @@ def describe_image(image, user_message):
     
     return qwq_result['choices'][0]['message']['content']
 
-# Function to handle the chat or image inputs
-def respond(message, history, system_message, max_tokens, temperature, top_p):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-    for message in qwq_client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message['choices'][0]['delta'].get('content', '')
-        response += token
-        yield response
-
-# Function to handle both image and chat input
-def chat_or_image(image, user_message, history, system_message, max_tokens, temperature, top_p):
+def chat_or_image(image, user_message):
     if image:
-        response = describe_image(image, user_message)
-        history.append((user_message, response))  # Keep history
-        return response, history
+        return describe_image(image, user_message)
     else:
-        return respond(user_message, history, system_message, max_tokens, temperature, top_p)
-
-# Gradio Blocks setup
-with gr.Blocks() as demo:
-    with gr.Row():
-        with gr.Column():
-            image_input = gr.Image(type="filepath", label="Upload image (Optional)")
-            user_input = gr.Textbox(label="Ask anything", placeholder="Ask...", lines=2)
-            system_message_input = gr.Textbox(value="You are a friendly assistant.", label="System message")
-            max_tokens_input = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens")
-            temperature_input = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
-            top_p_input = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)")
-            submit_button = gr.Button("Submit")
-
-        with gr.Column():
-            response_output = gr.Textbox(label="Response", lines=4, interactive=False)
-    
-    # Handle button click to trigger the callback
-    submit_button.click(
-        chat_or_image,
-        inputs=[
-            image_input,
-            user_input,
-            gr.State([]),  # History state to keep track of the conversation
-            system_message_input,
-            max_tokens_input,
-            temperature_input,
-            top_p_input
-        ],
-        outputs=[response_output, gr.State([])],  # Outputs: the response and updated history
-    )
+        qwq_result = qwq_client.chat_completion(
+            messages=[{"role": "user", "content": user_message}],
+            max_tokens=512,
+            temperature=0.7,
+            top_p=0.95
+        )
+        return qwq_result['choices'][0]['message']['content']
+
+demo = gr.Interface(
+    fn=chat_or_image,
+    inputs=[
+        gr.Image(type="filepath", label="Upload image (Optional)"),
+        gr.Textbox(label="Ask anything", placeholder="Ask...", lines=2)
+    ],
+    outputs="text",
+)
 
 if __name__ == "__main__":
-    demo.launch(show_error=True)
+    demo.launch(show_error=True)