upd

app.py

@@ -20,9 +20,7 @@ image_processor = git_processor_base.image_processor
 
 
 def preprocess_image(image):
-    """Преобразуем изображение для модели"""
-    image_rgb = image.convert("RGB")
-    return image_processor([image_rgb]).pixel_values
+    return git_processor_base(images=image, return_tensors="pt").pixel_values
 
 
 def translate_ru2en(text):
@@ -45,7 +43,7 @@ def generate_answer_git(image, question):
         "document-question-answering",
         model="andgrt/layoutlmv2-base-uncased_finetuned_docvqa",
     )
-    return qa_pipeline(image, question)[0]["answer"]
+    return qa_pipeline(preprocess_image(image), question)[0]["answer"]
 
 
 def generate_answer(image, question):