replace with distill llm

app.py

@@ -36,9 +36,9 @@ print("MPS available: ", torch.backends.mps.is_available())
 
 
 tokenizer = AutoTokenizer.from_pretrained(
-    "deepseek-ai/deepseek-coder-1.3b-instruct", trust_remote_code=True)
+    "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", trust_remote_code=True)
 model = AutoModelForCausalLM.from_pretrained(
-    "deepseek-ai/deepseek-coder-1.3b-instruct", 
+    "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", 
     trust_remote_code=True, 
     torch_dtype=torch.float16,  # Use float16 for better GPU memory efficiency
     device_map="auto"  # Automatically handle device placement