welcome to the new rabbitOS. we’ve upgraded the AI agent experience in rabbithole to tackle more complex tasks — with more thoughtful execution than ever before.
this is a free trial that’s available to anyone and everyone! whether you have an r1 or not yet, you can go to rabbithole and try it out today.
what is rabbitOS intern?
rabbitOS intern is our upgraded AI-native operating system that co-ordinates multiple agents to get things done! it currently features independent planning, research and coding abilities and is available for anyone to try for free on the rabbithole website, for a limited time.
users can assign rabbitOS intern tasks they might typically delegate to an intern at work, such as designing an interactive website for specific presentation use, conducting a comprehensive competitive analysis, building unique web apps and games, and much more. this version of intern is simply the first step - with many more features and optimisations planned, so we welcome all feedback!
to learn more, click here - we’re looking forward to hearing about your experiences!
But what would be the icing on the cake is if you could also change the language in the Rabbithole or if the Rabbithole would automatically orient itself to the language set in r1.
Congratulations to the entire Rabbit Inc. team for the outstanding work and impressive results achieved with rabbit r1… This release truly represents a leap forward in AI-powered devices, simplifying tasks and redefining user interaction. Thank you for your dedication and innovation… words alone cannot express our appreciation. Keep up the great work!
This way is very like as blackbox AI… truly appreciated!
This is an incredible piece of work. I’m so impressed. I had it do some searching for houses in my area that are selling. I gave it my address and told it the square feet, and what it returned was amazing. Is this going to be a subscription service from the talk about it?
It will likely have to be, although I believe r1 owners will get a better deal. It’s a very expensive service to run due to how it works, and we can’t afford to eat those costs as a business!
The exact economics of it are still being worked out, so until that happens, the free trial will continue.
To help stop discussions about r1 vs. subscriptions, I wrote the following summary to make a clear difference between the r1 and rabbitOS. Is this correct?
===
Rabbit has upgraded the AI agent experience (currently in the rabbithole) with a service that can tackle more complex tasks — with more thoughtful execution than ever before.
You can see the r1 and LAM Playground as a single AI agent that you control. rabbitOS is a multi AI agent. Although rabbitOS can currently be accessed in the rabbithole, it should be seen as separate environment. The only thing the two products have in common is:
rabbitOS reuses parts of rabbit’s technology
it can be accessed through the rabbithole
it is a product of rabbit
Owning an r1 is not a prerequisite for using rabbitOS. RabbitOS is a currently offered as a free trial to everyone, whether you have an r1 or not (yet). You can go to the rabbithole and try it out today.
Subscription in the making: The rabbitOS is a very expensive service to run due to how it works, and that is why there will be a subscription in the future (possibly when rabbitOS 2.0 is introduced ??). Until the subscriptions are introduced, use of rabbitOS will have some limitations in the number of prompts / chats that can be submitted. r1 holders have 9 chats per day. Everyone else gets 3 new chats per day. The exact economics of it are still being worked out, so until that happens, the free trial will continue.
To emphasize, rabbitOS is a new product (a multi AI agent) that is separate from the r1. The vision behind the r1 – the single AI agent – hasn’t changed. The r1 has no subscription costs.
So I was trying out the Intern feature in the recent days and I have to conclude that it is as useless as most AI Agents.
Reason:
If the User has limited Knowledge, it appears that AI Agents, or specifically the Intern, is doing a great job based on its given instructions. However when your own knowledge surpasses surface level, you start seeing incobsistencies or straight up not working samples.
In my Test the Agent four consecutive times on a simple task, however it was under rhe impression that it did everything right, and as every other AI agent that refuses to work, gave me instructions to solve the issue myself, instead of doing what I told it to do.
Four is a sample size too small, so I will continue testing and come back with further results.
I was hyped at first, but sadly I was not convinced yet by the intern.
Really cool! 