Quick summary
A custom GPT lets you build a personalized AI assistant in ChatGPT by defining its name, instructions, knowledge, and capabilities. This step-by-step walkthrough covers everything from opening the GPT builder to configuring access settings and publishing your finished model.
Steps
- On ChatGPT's main interface, open the left-side panel and click 'Explore GPTs'.
- Click the 'Create' button at the top-right corner to start building a new GPT.
- Click the 'Configure' tab to access the manual setup options.
- Click the 'Add' button under the profile photo section and upload an image to set a display photo for your custom GPT.
- Enter a unique Name and Description into their respective input fields.
- Write detailed guidelines covering behavior, purpose, personality, and capabilities, then type them into the 'Instructions' field.
- Add sample text prompts one by one into the 'Conversation Starters' input field.
- Click 'Upload Files' under the Knowledge section to attach external reference documents.
- Under Capabilities, tick the checkboxes for functions you want to enable, and optionally configure Actions to integrate with third-party tools.
- In the Preview Area, enter a test prompt and click 'Submit' to try your custom GPT.
- If satisfied with the result, click 'Create' to proceed.
- Configure your Access Settings based on your sharing preference, then click 'Save'.
- Your custom GPT is now created and published and ready to use.
