Update app.py

app.py

@@ -23,7 +23,7 @@ class CustomHuggingFaceLLM(LLM):
             llm_int8_enable_fp32_cpu_offload=True  # Offload FP32 operations to CPU for further memory savings
         )
         
-        self.model = AutoModelForCausalLM.from_pretrained(model_name,quantization_config=quantization_config)
+        self.model = AutoModelForCausalLM.from_pretrained(model_name, device_map="cpu", quantization_config=quantization_config)
         self.tokenizer = AutoTokenizer.from_pretrained(model_name)
         self.temperature = temperature