Update app.py

app.py

@@ -1,14 +1,23 @@
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
+import os
+
+
+# 获取环境变量中的 token
+hf_token = os.getenv("HF_TOKEN")
 
 # 加载模型和分词器
 model_path = "meta-llama/Llama-3.2-1B-Instruct-QLORA_INT4_EO8"
-tokenizer = AutoTokenizer.from_pretrained(model_path)
+tokenizer = AutoTokenizer.from_pretrained(
+    model_path,
+    token=hf_token
+)
 model = AutoModelForCausalLM.from_pretrained(
     model_path,
-    torch_dtype=torch.float16,
+    token=hf_token,
     device_map="auto",
+    load_in_4bit=True,  # 启用4-bit量化加载
 )
 
 def generate_text(prompt, max_length=512, temperature=0.7, top_p=0.9):