Update app.py

app.py

@@ -2,7 +2,7 @@ import torch
 from PIL import Image
 import gradio as gr
 import spaces
-from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer,BitsAndBytesConfig
+from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 import os
 from threading import Thread
 
@@ -12,7 +12,7 @@ from pptx import Presentation
 
 
 MODEL_LIST = ["nikravan/glm-4vq"]
-#MODEL_LIST = ["../Model_4b_sharded"]
+
 HF_TOKEN = os.environ.get("HF_TOKEN", None)
 MODEL_ID = MODEL_LIST[0]
 MODEL_NAME = "GLM-4vq"
@@ -32,19 +32,12 @@ h1 {
     display: block;
 }
 """
-inference_dtype=torch.bfloat16
-quantization_config = BitsAndBytesConfig(
-    load_in_4bit=True, bnb_4bit_compute_dtype=torch.float16
-)
 
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_ID,
-    torch_dtype=inference_dtype,
-    device_map = "cuda:0",
+    torch_dtype=torch.bfloat16,
     low_cpu_mem_usage=True,
-    trust_remote_code=True,
-    
-     quantization_config=quantization_config
+    trust_remote_code=True
 )
 tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)
 model.eval()