Update app.py

app.py

@@ -8,6 +8,7 @@ client = InferenceClient("meta-llama/Llama-3.2-3B-Instruct")
 
 
 def respond(
+    model_name,
     message,
     history: list[tuple[str, str]],
     system_message,
@@ -27,17 +28,25 @@ def respond(
 
     response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    # Model selection based on the button click
+    if model_name == "Llama":
+        for message in client.chat_completion(
+            messages,
+            max_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p,
+        ):
+            token = message.choices[0].delta.content
+            response += token
+    elif model_name == "Chatgpt":
+        response = "ChatGPT functionality is not yet implemented."
+    elif model_name == "Claude":
+        response = "Claude functionality is not yet implemented."
+    else:
+        response = "Model not recognized."
 
-        response += token
-        yield response
+    return response
 
 
 # CSS for styling the interface
@@ -61,27 +70,16 @@ body {
 }
 """
 
-# Function to trigger different AI models based on button click
-def trigger_model(model_name, message, history, system_message, max_tokens, temperature, top_p):
-    if model_name == "Llama":
-        # Here, you can choose the llama model to generate a response
-        return respond(message, history, system_message, max_tokens, temperature, top_p)
-    elif model_name == "Chatgpt":
-        # Placeholder for ChatGPT function (if needed)
-        return "ChatGPT response goes here."
-    elif model_name == "Claude":
-        # Placeholder for Claude function (if needed)
-        return "Claude response goes here."
-    else:
-        return "Model not found."
+# Define the Gradio interface with buttons and model selection
+def gradio_interface(model_name, message, history, system_message, max_tokens, temperature, top_p):
+    return respond(model_name, message, history, system_message, max_tokens, temperature, top_p)
 
 
-# Define the Gradio interface
 demo = gr.Interface(
-    fn=trigger_model,
+    fn=gradio_interface,
     inputs=[
         gr.Textbox(value="Hello!", label="User Message"),
-        gr.Textbox(value="System message", label="System Message", visible=False),
+        gr.Textbox(value="You are a virtual health assistant. Your primary goal is to assist with health-related queries.", label="System Message", visible=False),
         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(