Updated with OCR model and Gradio integration

app.py

@@ -1,34 +1,10 @@
 import gradio as gr
-from huggingface_hub import hf_hub_download
-from transformers import AutoTokenizer
-import importlib.util
-import sys
+from transformers import AutoModelForVision2Seq, AutoTokenizer
 
-# Define the repository name
-repo_id = "OpenGVLab/InternVL2-1B"
-
-# Download the correct files by inspecting the repository
-config_file = hf_hub_download(repo_id=repo_id, filename="configuration_internvl_chat.py", local_dir="model_files")
-model_file = hf_hub_download(repo_id=repo_id, filename="modeling_internvl_chat.py", local_dir="model_files")
-
-# Add the model_files directory to sys.path to enable absolute imports
-sys.path.append("model_files")
-
-# Load the configuration file dynamically
-spec_config = importlib.util.spec_from_file_location("InternVLChatConfig", config_file)
-InternVLChatConfig = importlib.util.module_from_spec(spec_config)
-spec_config.loader.exec_module(InternVLChatConfig)
-
-# Load the model file dynamically
-spec_model = importlib.util.spec_from_file_location("InternVLForVision2Seq", model_file)
-InternVLForVision2Seq = importlib.util.module_from_spec(spec_model)
-spec_model.loader.exec_module(InternVLForVision2Seq)
-
-# Load the tokenizer and model
+# Load the Hugging Face model and tokenizer, trusting remote code
 model_name = "OpenGVLab/InternVL2-1B"
 tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
-config = InternVLChatConfig.InternVLChatConfig.from_pretrained(model_name, trust_remote_code=True)
-model = InternVLForVision2Seq.InternVLForVision2Seq.from_pretrained(model_name, config=config, trust_remote_code=True)
+model = AutoModelForVision2Seq.from_pretrained(model_name, trust_remote_code=True)
 
 # Function to process and describe the image
 def analyze_image(image):