Update app.py

app.py

@@ -9,7 +9,7 @@ MODEL_FILE = "llama-3.2-1b-instruct.F16.gguf"
 # Adjust context window and other params as needed
 CONTEXT_WINDOW = 4096 
 MAX_NEW_TOKENS = 512
-TEMPERATURE = 0.7 
+TEMPERATURE = 1.5 
 
 # --- Model Loading Function ---
 def load_llm():
@@ -28,6 +28,7 @@ def load_llm():
         n_ctx=CONTEXT_WINDOW,
         n_threads=2,
         verbose=False # Set to True for debugging
+        min_p=0.1
     )
     print("Model loaded successfully!")
     return llm