feat: use GLM-4-flash model instead of GPT

app.py

@@ -28,8 +28,8 @@ from run_synthesis import (
 
 # API 配置从环境变量读取（通过 HF Secrets 设置）
 API_KEY = os.getenv("OPENAI_API_KEY")
-BASE_URL = os.getenv("OPENAI_BASE_URL", "https://api.openai.com/v1")
-DEFAULT_MODEL = "gpt-4o"
+BASE_URL = os.getenv("OPENAI_BASE_URL", "https://open.bigmodel.cn/api/paas/v4")
+DEFAULT_MODEL = "glm-4-flash"
 
 
 async def call_api(prompt: str, model: str = DEFAULT_MODEL, temperature: float = 0.7) -> str:
@@ -190,8 +190,8 @@ with gr.Blocks(title="UltraData-Math L3 Generator", css=custom_css) as demo:
     
     with gr.Row():
         model_select = gr.Dropdown(
-            choices=["gpt-4o", "gpt-4o-mini", "gpt-4-turbo", "gpt-3.5-turbo"],
-            value="gpt-4o",
+            choices=["glm-4-flash", "glm-4-plus", "glm-4-air", "glm-4-long"],
+            value="glm-4-flash",
             label="Model",
             scale=1,
         )