Update app.py

app.py

@@ -3,7 +3,8 @@ import gradio as gr
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model = AutoModelForCausalLM.from_pretrained("cyberagent/open-calm-7b", device_map="auto", torch_dtype=torch.float16)
+model = AutoModelForCausalLM.from_pretrained("cyberagent/open-calm-7b", device_map="auto", torch_dtype=torch.int8, load_in_8bit=True)
+                                                                                                            #torch_dtype=torch.float16)
 tokenizer = AutoTokenizer.from_pretrained("cyberagent/open-calm-7b")
 
 def proc( inputs ):