Update app.py

app.py

@@ -170,7 +170,7 @@ TOOLS = [
     ),
 ]
 
-MODEL_NAME = "openai/gpt-oss-20b"
+MODEL_NAME = "meta-llama/Llama-3.2-3B-Instruct"
 model = None
 tokenizer = None
 model_loaded = False
@@ -185,6 +185,9 @@ def download_and_load_model(progress=gr.Progress()):
         progress(0.2, desc="Downloading tokenizer...")
         tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
         
+        if tokenizer.pad_token is None:
+            tokenizer.pad_token = tokenizer.eos_token
+        
         progress(0.4, desc="Downloading model (this may take several minutes)...")
         model = AutoModelForCausalLM.from_pretrained(
             MODEL_NAME,
@@ -197,10 +200,10 @@ def download_and_load_model(progress=gr.Progress()):
         model_loaded = True
         
         progress(1.0, desc="Model loaded successfully!")
-        return "Model loaded successfully! You can now run queries."
+        return f"Model '{MODEL_NAME}' loaded successfully! You can now run queries."
     
     except Exception as e:
-        return f"Error loading model: {str(e)}\n\nNote: This model requires significant resources. Consider using a smaller model or API access."
+        return f"Error loading model: {str(e)}"
 
 def get_tool_descriptions() -> str:
     descriptions = []
@@ -455,7 +458,7 @@ with gr.Blocks(title="LLM Reasoning Modes Comparison") as demo:
     gr.Markdown("""
     # LLM Reasoning Modes Comparison
     
-    Compare three reasoning approaches using **openai/gpt-oss-20b**:
+    Compare three reasoning approaches using **Llama-3.2-3B-Instruct**:
     
     - **Think-Only**: Chain-of-Thought reasoning only (no tools)
     - **Act-Only**: Tool use only (no explicit reasoning)