Update2

app.py

@@ -1,20 +1,24 @@
 import gradio as gr
 from transformers import pipeline
 
-# Load a model (use a suitable model available on Hugging Face)
-model_name = "gpt2"  # Change to your desired model
+# Load a more suitable model for conversational responses
+model_name = "gpt2"  # You might want to try 'gpt-neo' or 'gpt-3.5-turbo' if available
 generator = pipeline("text-generation", model=model_name)
 
 # Inference function
-def generate_text(prompt):
-    return generator(prompt, max_length=50)[0]['generated_text']
+def generate_response(prompt):
+    # Generate text with a more structured approach
+    response = generator(prompt, max_length=100, num_return_sequences=1)[0]['generated_text']
+    return response.strip()  # Clean up any leading/trailing whitespace
 
 # Gradio interface
-interface = gr.Interface(fn=generate_text, 
-                         inputs="text", 
-                         outputs="text", 
-                         title="Text Generation with GPT-2",
-                         description="Enter a prompt to generate text.")
+interface = gr.Interface(
+    fn=generate_response, 
+    inputs="text", 
+    outputs="text", 
+    title="Conversational LLM",
+    description="Enter a message to receive a relevant response."
+)
 
 # Launch the interface
 interface.launch()