I’m having issues trying to get ChatGPT to Generate an Image for me.
Describe the problem/error/question
I’m trying to generate an Image using Dalle3 as part of a workflow, but I seem to run into an error. It works when i use Dalle2, just not when I swap it to Dalle3.
What is the error message (if any)?
The error I’m getting from N8N is: The service is receiving too many requests from you Error in request. Please check your input.
The error details are: Error code429
Full message
{ "error": { "message": "Error in request. Please check your input.", "type": "invalid_request_error", "param": null, "code": null } }
Please share your workflow
It’s a pretty big workflow, so I’ll just copy the snippet
Did you check the requests per timeframe you can send to each one?
I think the error message is quite obvious.
Usually the older / less complex models are cheaper and/or offer more requests for free. (Or maybe the one that you are trying doesn’t have a free tier and is payable)
Hi, I don’t know the exact reason. But 429 is too many request to the API. I guess you could find out from the executions data / view on which node it stops and most likely that node call would be responsible.
I think the easiest after that is to go into the openai API control panel and look into your usage. It’s quite granular per model etc
Hi, this doesn’t say much really. The most important ones are the request per minute imho. Don’t remember how it shows exactly but I think there is a threshold line somewhere.
I don’t know the exact limits since they are based on several factors. But here’s a public statement of something similar
You can find and set / monitor limits in your account. There should be a limits section on your account where you can see your actual limits per your defined project.
Btw I checked mine and I’m at only 5 dalle-3 IMG per min max so if you are on the same you need to add wait nodes into your workflow when automated or space your tests
It doesn’t look like I’ve set any rate limits, the only things there, are what was default.
Do you think I should raise the limits?
I also looked at budget limits, which I set for about 100/month, I don’t think this would stop it.
As for the default limit, I haven’t had a single successful image generated for dalle-3 (confirmed as it doesn’t show up on the openAI usage dashboard).
Hi, the only thing I can think of is try with an external tool and validate whether it’s the same error… Or another test to use the http request or an external tool to get back the info on what openai thinks you have still left. It’s not an issue of N8n according to me
Maybe creating another project with an additional API key might kick some sense into openAI
