Update app.py

app.py

@@ -1,20 +1,17 @@
 import gradio as gr
 from functools import lru_cache
-import openai  # 用于调用外部API
 import os
 import spaces
 import gradio as gr
 from transformers import AutoTokenizer, AutoModel,AutoModelForCausalLM
-import platform
 import torch
-import nltk
-from functools import lru_cache
 
 # 假设openai_client已定义，例如：
 
 device = "cuda"
 MODEL_NAME = "ByteDance-Seed/Seed-X-PPO-7B"
 
+print("Start dowload")
 def load_model():
     model = AutoModelForCausalLM.from_pretrained(MODEL_NAME,torch_dtype="bfloat16").to(device)
     print(f"Model loaded in {device}")
@@ -22,7 +19,7 @@ def load_model():
 
 
 model = load_model()
-
+print("Ednd dowload")
 # Loading the tokenizer once, because re-loading it takes about 1.5 seconds each time
 tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
 
@@ -88,4 +85,6 @@ with gr.Blocks() as demo:
         outputs=output,
     )
     examples = gr.Examples(examples=examples_inputs,inputs=[input_text], fn=translate, outputs=output, cache_examples=True)
+
+print("Prepared")
 demo.launch()