fix: workaround for ORTModel that cant't access private repository even with an auth token

app.py

@@ -6,6 +6,8 @@ import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 from optimum.onnxruntime import ORTModelForCausalLM
 
+from huggingface_hub import login
+
 import gradio as gr
 
 MODEL_NAME = (
@@ -15,26 +17,24 @@ MODEL_NAME = (
 )
 HF_READ_TOKEN = os.environ.get("HF_READ_TOKEN")
 
+login(HF_READ_TOKEN)
+
 assert isinstance(MODEL_NAME, str)
 
 tokenizer = AutoTokenizer.from_pretrained(
     MODEL_NAME,
     trust_remote_code=True,
-    token=HF_READ_TOKEN,
 )
 model = {
     "default": AutoModelForCausalLM.from_pretrained(
         MODEL_NAME,
-        token=HF_READ_TOKEN,
     ),
     "ort": ORTModelForCausalLM.from_pretrained(
         MODEL_NAME,
-        use_auth_token=HF_READ_TOKEN,
     ),
     "ort_qantized": ORTModelForCausalLM.from_pretrained(
         MODEL_NAME,
         file_name="model_quantized.onnx",
-        use_auth_token=HF_READ_TOKEN,
     ),
 }
 
@@ -676,7 +676,7 @@ def demo():
             )
 
     ui.launch(
-        share=True,
+        # share=True,
     )