Update app.py

app.py

@@ -37,9 +37,8 @@ recommended_threads = 4
 # === Load the model ===
 llm = Llama(
     model_path=model_path,
-    n_ctx=8192,  # Can increase depending on memory
+    n_ctx=2048,  # Can increase depending on memory
     n_threads=recommended_threads,
-    n_batch=32,  # adjust depending on RAM
     use_mlock=True,  # lock model in RAM for faster access
     n_gpu_layers=0,  # CPU only, use >0 if GPU is present
     chat_format="chatml",  # for Hermes 2