An error occurred. Please try again later. Click to see more details.
Agent: page
Name: ValidationException
Message: The model returned the following errors: Input is too long for requested model.
Yeah i get the Same Error.
But it does not seem that long tho
Hello guys, a fix is in QA for the “Page too large”, “Input too long” issue. We hope to release it asap this month.
Hey, @DanielL just thought I’d chime in since it has been a few weeks since QA, but this my attempt today. Happy to share more details or the project if it helps you.
@DanielL Do you have an ETA on this fix? We have not been able to use any AI features for months on any element because of this issue.
I have been requesting an update on the AI builder for some weeks now. I am paying for AI tokens but cannot use them. We need a fix for this or at least to transfer this month’s AI tokens to next month. Otherwise i think a refund for the AI tokens is appropriate.
Hello everyone, just wanted to give you an update on this ongoing issue.
The error you’re seeing comes from the way our AI feature currently handles context.
Right now, it often tries to send the entire app as context even when you pass it a specific element.
As a result, the request often goes beyond what the model accepts.
The tech team has been actively working on an update that will give you better control over what goes into the context and should help avoid the issue in most cases. Unfortunately, we still haven’t landed on a solution that’s solid enough to push to production.
There’s been some back and forth between development and QA, and while things are moving forward, we’re not quite at the point where we can share an ETA with confidence.
On the billing side: Because AI tokens are bundled with your WeWeb subscription, we don’t offer partial refunds.
But if you’d rather pause your subscription until the fix is ready, you can cancel your seat plan(s) anytime and we’ll make sure your last payment is refunded.
We’ll keep you posted here as soon as the update goes live.
So how far are we or how close to the solution?
Hello everyone , we just released a couple features to help with this issue!
You can read more about all changes for this release here: