Update app.py

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import spaces
 import torch
+import os
 print(f"Is CUDA available: {torch.cuda.is_available()}")
 # True
 print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
@@ -11,7 +12,7 @@ print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
 # client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-client = InferenceClient("deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B")
+client = InferenceClient("deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", token=os.getenv(hftoken))
 
 @spaces.GPU
 def respond(