Update app.py

app.py

@@ -1,12 +1,13 @@
 import gradio as gr
 import torch
-import torchvision
+#import torchvision
 import base64
 import fitz  # PyMuPDF
 import tempfile
 from io import BytesIO
 from PIL import Image
-from transformers import AutoProcessor, AutoModelForVision2Seq
+#from transformers import AutoProcessor, AutoModelForVision2Seq
+from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration
 
 from olmocr.data.renderpdf import render_pdf_to_base64png
 from olmocr.prompts.anchor import get_anchor_text
@@ -15,6 +16,16 @@ import re
 import html
 import json
 
+
+#from olmocr.data.renderpdf import render_pdf_to_base64png
+#from olmocr.prompts import build_no_anchoring_v4_yaml_prompt
+
+# Initialize the model
+model = Qwen2_5_VLForConditionalGeneration.from_pretrained("allenai/olmOCR-2-7B-1025", torch_dtype=torch.bfloat16).eval()
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct")
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
 #processor = AutoProcessor.from_pretrained("allenai/olmOCR-2-7B-1025-FP8")
 #model = AutoModelForVision2Seq.from_pretrained("allenai/olmOCR-2-7B-1025-FP8",
 #            torch_dtype=torch.bfloat16,
@@ -22,10 +33,10 @@ import json
 #            low_cpu_mem_usage=True
 #        )
 
-model = AutoModelForVision2Seq.from_pretrained("allenai/olmOCR-2-7B-1025-FP8", torch_dtype=torch.bfloat16).eval()
-processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct")
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-model.to(device)
+#model = AutoModelForVision2Seq.from_pretrained("allenai/olmOCR-2-7B-1025-FP8", torch_dtype=torch.bfloat16).eval()
+#processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct")
+#device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+#model.to(device)
 
 
 #model = Qwen2VLForConditionalGeneration.from_pretrained(