Update src/model_load.py

src/model_load.py

@@ -14,7 +14,8 @@ def load_model():
     tokenizer = AutoTokenizer.from_pretrained(model_name)
     model = AutoModelForCausalLM.from_pretrained(model_name,
                                                  torch_dtype=torch.float16,
-                                                 device_map="auto")
+                                                 device_map="auto",
+                                                 quantization_config=BitsAndBytesConfig(load_in_8bit=True))
     
     # Crear pipeline de generación de texto
     text_generation_pipeline = pipeline(