Update app.py

app.py

@@ -12,16 +12,9 @@ model_path = hf_hub_download(
 
 llm = Llama(
     model_path=model_path,
-    #n_ctx=2048,
-    #n_threads=8,
+    n_ctx=2048,
     chat_format="llama-3",
-    #seed=42,
-    #f16_kv=True,
-    #logits_all=False,
-    #use_mmap=True,
-    #use_gpu=True,
-    #n_gpu_layers=-1,  # to ensure all layers are on GPU
-    #offload_kqv=True  # for better memory management    
+    n_gpu_layers=-1,  # ensure all layers are on GPU
 )
 
 # Placeholder responses for when context is empty