I got my rabbit R1 about 2 weeks ago, and I am super excited about the potential applications that can be built with it. I am incredibly impressed with the hardware; both its retro and simple design.
It would be tremendously wise of the Rabbit team to develop and release an API for users. More specifically, giving users the ability to access input utterances and to send back responses to the R1. Having this would accelerate the development of disruptive applications on the R1, by expanding the developmental features to external developers and having a platform for users to install custom applications (i.e. an “App Store”).
Again, super excited to see the promise of the R1 to transform handheld technology.
Thanks for sharing your thoughts on the Rabbit R1. I’ve had mine for about 3 days now, and I completely agree with your excitement about its potential. The hardware is indeed impressive, with its retro and simple design that’s both functional and aesthetically pleasing.
Your point about developing and releasing an API is spot on, and I couldn’t agree more. As someone who’s also into programming AI agents, I can see tremendous value in this. Currently, I’m using Telegram for input because it allows me to send audio, text, and pictures. An API for the Rabbit R1 would be a game-changer, allowing us to easily integrate it into our existing setups or create entirely new applications.
Building on your idea, I think the ideal API would include the following:
Inputs:
Text
Audio
Photo
Video
Outputs:
Text
Photo
Video
Audio
Text to Audio (using Rabbit’s native voice)
Having these input and output options would open up so many possibilities. Imagine being able to trigger custom calls on the device or use Rabbit’s native voice for audio output. This would allow for seamless integration of Rabbit’s responses into our applications.
Another crucial feature would be mapping the output in Rabbit Hole. This way, when RAG (Retrieval-Augmented Generation) is implemented and follow-up questions become possible, that knowledge would be stored in the system for future requests. This would greatly enhance the device’s learning capabilities and make it more useful over time.
I also think your suggestion of a platform for users to install custom scenarios is brilliant. This would create a thriving ecosystem around the R1, because the ease of use and openness for integration would set it apart from all other Devices. On the phone for example you needa use Telegram for example or always add on more Apps. On the Rabbit that all could be native and just as easy to use as the standard features.
The potential for the R1 to transform handheld technology is indeed exciting. With an open API and a supportive developer community, we could see some truly disruptive applications. It would be great to hear from the Rabbit team about their plans for developer support and API access.
Has anyone else here been experimenting with AI agents or have ideas for potential R1 applications? I’d love to hear your thoughts and experiences!