Update llm.py

llm.py

@@ -1,12 +1,13 @@
-import torch
+import os
+hf_token=os.getenv("HUGGINGFACE_HUB_TOKEN")import torch
 import asyncio
 from functools import partial
 import time
 from transformers import AutoProcessor, AutoModelForImageTextToText #, BitsAndBytesConfig
-import os
+
 # Quantization config
 # quant_config = BitsAndBytesConfig(load_in_8bit=True)
-hf_token=os.getenv("HUGGINGFACE_HUB_TOKEN")
+
 # Load processor
 processor = AutoProcessor.from_pretrained("datalab-to/chandra-ocr-2")