Update app.py

app.py

@@ -7,7 +7,7 @@ device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
 
 b_tokenizer = AutoTokenizer.from_pretrained("bigscience/bloom-7b1")#using small parameter version of model for faster inference on hf
-b_model = AutoModelForCausalLM.from_pretrained("bigscience/bloom-7b1")
+b_model = AutoModelForCausalLM.from_pretrained("bigscience/bloom-7b1",device_map = "auto")
 
 g_tokenizer = AutoTokenizer.from_pretrained("google/gemma-2-2b",token = hf_token)#using small paramerter version of model for faster inference on hf
 g_model = AutoModelForCausalLM.from_pretrained("google/gemma-2-2b",token = hf_token)