Update app.py

app.py

@@ -1,4 +1,5 @@
 import gradio as gr
+import transformers
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from transformers.generation import GenerationConfig
 import re
@@ -15,22 +16,15 @@ from PIL import Image, ImageDraw, UnidentifiedImageError
 
 base_url = "https://huggingface.co/spaces/Tonic1/Official-Qwen-VL-Chat"
 model_name = "Qwen/Qwen-VL-Chat"
-tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", fp16=True, trust_remote_code=True).eval()
-model.generation_config = GenerationConfig.from_pretrained(model_name, trust_remote_code=True)
-
-device = "cuda" if torch.cuda.is_available() else "cpu"
-model.to(device)
-task_history = []
 
 BOX_TAG_PATTERN = r"<box>([\s\S]*?)</box>"
 PUNCTUATION = "！？。＂＃＄％＆＇（）＊＋，－／：；＜＝＞＠［＼］＾＿｀｛｜｝～｟｠｢｣､、〃》「」『』&#8203;``【oaicite:0】``&#8203;〔〕〖〗〘〙〚〛〜〝〞〟〰〾〿–—‘’‛“”„‟…‧﹏."
 
 class ChatBot:
     def __init__(self):
-        self.tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen-VL-Chat", trust_remote_code=True)
-        self.model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen-VL-Chat", device_map="cuda", trust_remote_code=True).eval()
-        self.model.generation_config = GenerationConfig.from_pretrained("Qwen/Qwen-VL-Chat", trust_remote_code=True)
+        self.tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+        self.model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", bf16=True, trust_remote_code=True).eval()
+        self.model.generation_config = GenerationConfig.from_pretrained(model_name, trust_remote_code=True)
         self.history = []
 
     def chat(self, image_path=None, text_query=None):