diff --git a/app_chatgpt/data/ai_robot_data.xml b/app_chatgpt/data/ai_robot_data.xml
index 463599d3..a3ecb311 100644
--- a/app_chatgpt/data/ai_robot_data.xml
+++ b/app_chatgpt/data/ai_robot_data.xml
@@ -26,7 +26,7 @@
         <field name="provider">azure</field>
         <field name="endpoint">https://my.openai.azure.com</field>
         <field name="engine">gpt35</field>
-        <field name="api_version">2024-02-15-preview</field>
+        <field name="api_version">2024-03-01-preview</field>
         <field name="sequence">8</field>
         <field name="image_avatar" type="base64" file="app_chatgpt/static/description/src/chatgpt35_azure.png"/>
     </record>
@@ -36,7 +36,7 @@
         <field name="provider">azure</field>
         <field name="endpoint">https://my.openai.azure.com</field>
         <field name="engine">gpt4</field>
-        <field name="api_version">2024-02-15-preview</field>
+        <field name="api_version">2024-03-01-preview</field>
         <field name="sequence">9</field>
         <field name="image_avatar" type="base64" file="app_chatgpt/static/description/src/chatgpt4_azure.png"/>
     </record>
@@ -45,7 +45,7 @@
         <field name="provider">azure</field>
         <field name="endpoint">https://my.openai.azure.com</field>
         <field name="engine">gpt4-32k</field>
-        <field name="api_version">2024-02-15-preview</field>
+        <field name="api_version">2024-03-01-preview</field>
         <field name="sequence">10</field>
         <field name="image_avatar" type="base64" file="app_chatgpt/static/description/src/chatgpt4_azure.png"/>
     </record>
diff --git a/app_chatgpt/models/ai_robot.py b/app_chatgpt/models/ai_robot.py
index 649ae31b..54059edc 100644
--- a/app_chatgpt/models/ai_robot.py
+++ b/app_chatgpt/models/ai_robot.py
@@ -35,7 +35,7 @@ class AiRobot(models.Model):
         ('code-davinci-002', 'Chatgpt 2 Code Optimized'),
         ('text-davinci-002', 'Chatgpt 2 Davinci'),
         ('dall-e2', 'Dall-E Image'),
-    ], default='gpt-4',
+    ], default='gpt-3.5-turbo-0125',
                                 help="""
 GPT-4: Can understand Image, generate natural language or code.
 GPT-3.5: A set of models that improve on GPT-3 and can understand as well as generate natural language or code