Learning Timeline
Key Insights
Benefits of Manually Selecting a Model
Using the Model Picker gives you precise control over output quality. This is especially useful when building GPTs for colleagues, as it provides them with a powerful experience without requiring them to learn prompt engineering themselves.
OpenAI's 'Silent Update' Feature
The Model Picker was rolled out as a silent update without an official announcement. If you have older GPTs, make sure to revisit the Configure tab to update your model to the most appropriate version.
Step by Step
How to Set a Specific Model (Model Picker) in a Custom GPT
- Open ChatGPT and log in to your account.
- Click the 'Explore GPTs' menu in the left sidebar.
- Select 'Create' to build a new GPT, or click on an existing GPT and select 'Edit'.
- Click the 'Configure' tab at the top of the screen.
- Locate the newly available 'Model' section (Model Picker) in the configuration list.
- Click the dropdown menu to view the list of available OpenAI models.
- Select the specific model you want to use (e.g., GPT-4o or GPT-4) to ensure consistent output quality.
- Complete the 'Instructions' section with a prompt optimized for the specific task context.
- Fill in 'Conversation starters' to set up initial messages for users.
- Upload reference documents in the 'Knowledge' section by clicking the 'Upload files' button if necessary.
- Click the 'Save' or 'Update' button in the top-right corner of the screen.
- Select 'Only me', 'Anyone with a link', or 'Public' and click 'Confirm' to activate the model settings.