Kookee now has three plans: Free, Pro at $29 a month and Ultra at $99 a month, excluding tax. Every number on the new pricing page comes from the limits the product enforces, and what paid plans mainly buy is AI answers.
New
Free, Pro and Ultra. $0, $29 and $99 a month, excluding tax: Paddle, which handles payments, adds any tax due at checkout. On the pricing page, the upgrade buttons open your billing settings when you are signed in.
AI answers are the AI allowance. A reply from the assistant that links at least one of your articles counts as one answer: 50 a month on Free, 450 on Pro and 1,650 on Ultra, shared by all your projects. Greetings, replies that found nothing and errors are free. The billing tab, the dashboard, each project's overview and its card show answers used, in place of chat tokens.
No "powered by kookee.dev" link on Pro and Ultra. The portal footer shows only your project's name.
A limit of its own for feedback posts. A project holds up to 1,000 feedback posts on Free, 25,000 on Pro and 250,000 on Ultra, and they no longer count toward its entries.
Daily caps for visitors. Someone with no role in a project can post 10 feedback posts and 50 comments in it in any 24 hours, the same on every plan. Members of the project have no daily cap.
Changed
New limits.
Free | Pro | Ultra | |
|---|---|---|---|
Projects per account | 1 | 5 | 25 |
API requests a month | 100,000 | 2,000,000 | 10,000,000 |
API requests a minute per project | 300 | 1,000 | 5,000 |
AI answers a month | 50 | 450 | 1,650 |
AI embedding tokens a month | 2,000,000 | 20,000,000 | 50,000,000 |
Feedback posts per project | 1,000 | 25,000 | 250,000 |
Every other limit stays as it was. An account that already holds more than a new limit allows keeps everything, and can add more once it is back under the limit.
The chat keeps answering when embedding tokens run out. Embedding tokens pay for indexing your articles for AI search, so once they are used up, new and changed articles wait for the 1st of the month. The chat itself stops only at the month's answers or its AI cost ceiling.
Usage emails give the AI budget as a share. The email about the monthly AI budget says how much of it is used, as a percentage, rather than a token count.
The Site widget overview drops its Total Tokens tile. Answers Used shows the answers of all your projects against your plan's allowance.
Fixed
A usage warning is sent once. After the 100% warning for a limit, the hourly check no longer follows it with 90% and 80% ones.
Getting it
The new plans and limits apply to every account now. Prices are on the pricing page, and every limit with its error code is in Rate limits, plans and error handling.