AI startup rabbit inc. has launched a free trial of rabbitOS intern, an upgraded AI-native operating system designed to coordinate multiple AI agents to perform complex tasks. Featuring a general agency that can reason, plan, and execute code-level functions from user prompts, rabbitOS intern exhibits intern-level capabilities across multiple domains. Initially developed for the r1 device, rabbitOS intern will eventually support a broader range of rabbit products and integrate with various digital platforms and compatible devices.
The trial is available for free for a limited time. r1 owners receive nine task executions per day, while all other users get three.
Users can delegate diverse tasks to rabbitOS intern — much like assigning tasks to a real intern. Early users have seen impressive results including:
- Generating regional real estate research reports
- Developing a Three.js airplane game with aliens
- Creating interactive TV show websites
- Building a 16-bit music sequencer hosted on a website
- Performing financial analyses of companies
The growing demand for AI agents that demonstrate autonomy is reshaping human-computer interaction. rabbitOS intern represents a significant leap, shifting from simple automation to collaborative AI with planning and execution abilities. Much like a human intern gaining experience, rabbitOS intern is expected to evolve toward professional-level performance.
rabbit’s vision is to create a general AI agent that operates across multiple platforms — from web and desktop to mobile and future device formats. The latest version of rabbitOS intern integrates existing agent tools such as:
- LAM playground: a versatile web agent
- Teach mode: a teachable web agent
- Android agent: for mobile platforms
These will be further developed and merged into rabbitOS intern in future updates.
While rabbitOS intern will become a paid feature, it remains accessible to all during the trial period. The r1 device will continue to be subscription-free. rabbit inc. will use this period to gather feedback on user experience, feature preferences, and pricing strategies.
Note that during the trial phase, performance may vary depending on task complexity, and some errors may occur. Still, this open access period reflects rabbit's belief in the transformative potential of AI collaboration and human-like agency.