Trial and error.

app.py

@@ -34,7 +34,7 @@ class BasicAgent:
             print("Using Inference API for generation...")
             response = self.client.text_generation(
                 prompt=question,
-                model="meta-llama/Llama-3-8B-Instruct",  # Free model
+                model="google/gemma-7b-it",  # Free model
                 max_new_tokens=500,
                 temperature=0.7,
                 do_sample=True,