feat: Refactor chatbot functionality to support multiple models and improve user interaction with a new input system.

app.py

@@ -2,77 +2,101 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 
 MODELS = [
-    "openai/gpt-oss-20b",
-    "Qwen/Qwen3-30B-A3B-Instruct-2507",
+    "Qwen/Qwen2.5-72B-Instruct",
+    "meta-llama/Llama-3.1-8B-Instruct",
 ]
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    model_name,
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model=model_name)
+def user(user_message, history_a, history_b):
+    if not user_message:
+        return "", history_a, history_b
+    
+    user_msg = {"role": "user", "content": user_message}
+    
+    return "", history_a + [user_msg], history_b + [user_msg]
 
-    messages = [{"role": "system", "content": system_message}]
+def bot(history, model_name, system_message, max_tokens, temperature, top_p, oauth_token: gr.OAuthToken | None):
+    if not history:
+        yield history
+        return
+        
+    if oauth_token is None:
+        history.append({"role": "assistant", "content": "⚠️ Please login via the Sidebar."})
+        yield history
+        return
 
-    messages.extend(history)
+    client = InferenceClient(token=oauth_token.token, model=model_name)
+    
+    # Prepend system message for the API call
+    messages = [{"role": "system", "content": system_message}] + history
+    
+    # Initialize assistant message
+    history.append({"role": "assistant", "content": ""})
+    
+    try:
+        stream = client.chat_completion(
+            messages,
+            max_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p,
+        )
+        
+        response_text = ""
+        for chunk in stream:
+            token_content = chunk.choices[0].delta.content or ""
+            response_text += token_content
+            history[-1]["content"] = response_text
+            yield history
+            
+    except Exception as e:
+        history[-1]["content"] = f"**Error:** {str(e)}"
+        yield history
 
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
+with gr.Blocks() as demo:
+    with gr.Sidebar():
+        gr.Markdown("## Configuration")
+        gr.LoginButton() # 'scopes' argument removed in Gradio 6
+        system_message = gr.Textbox(value="You are a friendly Chatbot.", label="System message")
+        max_tokens = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens")
+        temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
+        top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p")
 
-        response += token
-        yield response
+    gr.Markdown("# Chatbot Arena (Gradio 6.x)")
 
+    with gr.Row():
+        with gr.Column():
+            model_a = gr.Dropdown(MODELS, label="Model A", value=MODELS[0])
+            chatbot_a = gr.Chatbot(label="Model A Output") # 'type' argument removed
+        
+        with gr.Column():
+            model_b = gr.Dropdown(MODELS, label="Model B", value=MODELS[1])
+            chatbot_b = gr.Chatbot(label="Model B Output")
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    type="messages",
-    additional_inputs=[
-        gr.Dropdown(
-            MODELS, label="Select Model", value=MODELS[0], interactive=True
-        ),
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+    with gr.Row():
+        msg = gr.Textbox(scale=4, label="Input Message", autofocus=True)
+        clear_btn = gr.Button("Clear Chat", scale=1)
 
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
+    # Event handlers
+    submit_event = msg.submit(
+        user, 
+        [msg, chatbot_a, chatbot_b], 
+        [msg, chatbot_a, chatbot_b], 
+        queue=False
+    )
+    
+    submit_event.then(
+        bot,
+        [chatbot_a, model_a, system_message, max_tokens, temperature, top_p],
+        [chatbot_a]
+    )
+    
+    submit_event.then(
+        bot,
+        [chatbot_b, model_b, system_message, max_tokens, temperature, top_p],
+        [chatbot_b]
+    )
 
+    clear_btn.click(lambda: ([], []), None, [chatbot_a, chatbot_b])
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(debug=True)