For OpenAI API issues: Check your API key at OpenAI API Keys and update it in Clay. Ensure your plan meets requirements: 30,000 TPM for ChatGPT columns and 450,000 TPM for Claygent.
For Anthropic API issues: Visit Anthropic API settings to check your limits. Lower tier accounts need to process data in smaller batches of around 50 rows.
If you get a 401 "invalid API key" error, verify you're using the correct API key in your Clay integration settings.
Never mind, I figured it out. Basically, you need to run the prompt generation on the model that you're going to run it on, not on Claygent (i.e. Argon/Neon)
