Switching Custom GPTs to Reasoning Models (o1/o3) for Better Results | Alpha | PandaiTech
PandaiTech Logo

Switching Custom GPTs to Reasoning Models (o1/o3) for Better Results

A guide to configuring Custom GPTs to use the latest models like o1 or o3 to solve complex coding and logic problems.

Learning Timeline
Key Insights

When to Use the o1/o3 Series

Use reasoning models (o1/o3) if your Custom GPT focuses on heavy-duty tasks such as solving mathematical problems, complex coding, or logic that requires deep reasoning, as opposed to GPT-4o which is optimized for speed.

User Flexibility

Even if you set a default model in the 'Configure' section, end users still have the option to manually switch models at the top of the chat screen if they require different results.
Step by Step

How to Switch Your Custom GPT Model to the Reasoning Series (o1/o3)

  1. Open ChatGPT and click on 'Explore GPTs' in the left sidebar.
  2. Click the 'Create' button to build a new GPT, or select an existing Custom GPT and click the 'Edit' (pencil) icon.
  3. Click on the 'Configure' tab located at the top next to the 'Create' tab.
  4. Locate the 'Recommended model' section, which indicates which model will be used by default for users.
  5. Click the dropdown menu under 'Recommended model'.
  6. Choose your preferred reasoning model from the list, such as 'o3', 'o3 Pro', or 'o1'.
  7. Click the 'Update' or 'Save' button in the top right corner of the screen to activate the changes.
  8. To switch models during an active chat session, click the model name at the top of the chat window and select an alternative model from the dropdown menu.