Troubleshooting Integration Error with GPT 4.1 Mini | Clay

Troubleshooting Integration Error with GPT 4.1 Mini

Clay S. why am i getting this error "The integration failed to run ❌ Invalid Request" when I run the prompt with a GPT 4.1 mini but the prompt runs if I run it with an O1. Keep in mind the prompt was working fine like three days ago

7 comments

Channeled
APP
·
·
This error may be related to your OpenAI API integration. The issue could be due to: 1. An incorrect or expired API key - verify your key at platform.openai.com/account/api-keys 2. Usage limits on your OpenAI account for this model 3. Changes in model requirements since it worked three days ago Try updating your API key in Clay and check if you've reached usage limits in your OpenAI account. If the problem persists, consider making your prompts more specific or breaking them into smaller requests.
Anthony U.
·
·
Clay S. I am having the same problem. This is seems to be an API or internal issue
Channeled
APP
·
·
This error is likely related to your OpenAI API integration. The issue could be due to: 1. An incorrect or expired API key - verify your key at platform.openai.com/account/api-keys 2. Usage limits on your OpenAI account for this specific model 3. Changes in model requirements since it worked three days ago For reliable use of OpenAI in Clay, you need sufficient tokens per minute (TPM). Check your API usage tier at platform.openai.com/settings/organization/limits to ensure it meets the requirements.
Bo (.
·
·
Hey - sorry about this, it looks like we've run into a bug with ChatGPT. I've escalated it to our engineering team, and we'll follow up as soon as we have more info. In the meantime, if it's urgent, try using a different provider (like Claude or Gemini) - it should work. We appreciate your patience.
Adam A.
·
·
oh man ok please keep me posted
Bo (.
·
·
Good news! We’ve heard back from the team, and the issue with ChatGPT errors should now be resolved. Go ahead and do a quick refresh (Cmd/Ctrl + Shift + R), then filter the column for rows with errors. Once filtered, click the column header for the ChatGPT column, select Run Column, and re-run just the rows that errored out. Thanks again for your patience while we worked through this! Let me know if you have more questions.
Channeled
APP
·
·