Update app.py

app.py

@@ -71,13 +71,11 @@ def retrieve_context(query):
 # Load Qwen model (CPU only, no accelerate)
 # -----------------------------
 model_name = "meta-llama/Llama-3.2-1B"
-
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
-    torch_dtype=torch.float32  # CPU only
+    torch_dtype=torch.float32  # CPU-friendly
 )
-
 generator = pipeline(
     "text-generation",
     model=model,
@@ -85,7 +83,7 @@ generator = pipeline(
     max_new_tokens=150,
     do_sample=True,
     temperature=0.6,
-    device=-1  # ensures CPU is used
+    device=-1  # CPU
 )
 
 print("LLM loaded successfully!")