Trial and error.

app.py

@@ -38,6 +38,8 @@ class BasicAgent:
 
             # Generate response
             if self.model and self.tokenizer:
+                printf("Using local model for generation...")
+
                 # Local model generation
                 text = self.tokenizer.apply_chat_template(
                     messages,
@@ -60,6 +62,8 @@ class BasicAgent:
 
                 answer = self.tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
             else:
+                print("Using Inference API for generation...")
+
                 # Fallback to Inference API
                 response = self.client.chat(messages=messages)
                 answer = response.content
@@ -198,7 +202,7 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
 
 # --- Build Gradio Interface using Blocks ---
 with gr.Blocks() as demo:
-    gr.Markdown("# Basic Agent Evaluation Runner #1")
+    gr.Markdown("# Basic Agent Evaluation Runner #2")
     gr.Markdown(
         """
         **Instructions:**