For the current Analyze and Modify Airtable Data with ChatGPT automation, you can set a maximum token allowance for the response. You can find this in:
Your automation's configuration page
Advanced
Set Completion Max Tokens (Optional)
When this setting is blank, miniExtensions requests a maximum of 4,000 tokens. Values greater than 0 and no greater than 4,000 are used; other values fall back to 4,000. This limit still applies when you use your own OpenAI API key. The actual response may be shorter and is subject to the selected model's limits.
The legacy Analyze and Modify Airtable Data using the GPT Model automation behaves differently: when you use your own OpenAI API key, it ignores this setting and omits the miniExtensions completion-token limit from the request. Without your own key, only Test Mode is available, and it uses the same 4,000-token default and valid range described above. OpenAI and model limits still apply.