increase ctx lenght to 2k

app.py

@@ -113,7 +113,7 @@ def try_load_model(path):
     try:
         return Llama(
             model_path=path,
-            n_ctx=512,  # Reduced context window
+            n_ctx=2048,  # Reduced context window
             n_threads=2,
             n_threads_batch=1,
             n_batch=64,