thesysdev/appless

What if your phone had no apps

What it solves

AppLess reimagines the mobile user experience by replacing traditional, static apps and menus with a generative UI operating system. Instead of navigating through pre-installed icons and fixed interfaces, users simply ask for what they need, and the system generates a custom, native-looking screen on the fly.

How it works

The system uses a component contract (ui/contract.tsx) as a single source of truth to define the UI vocabulary available to the model. This contract generates a system prompt for an LLM (via Cerebras), which then streams back a compact UI language called openui-lang. An OpenUI renderer parses this stream incrementally and converts it into real native components—using Cupertino for iOS and Material 3 for Android—so the interface appears live before the model finishes responding. To provide real-world data, the model can call tools like Exa for web search mid-stream.

Who it’s for

It is primarily an experimental project for developers and designers interested in Generative UI and those who want to build a "app-less" OS experience by integrating real APIs into the simulated framework.

Highlights

  • Generative Native UI: Renders real native components on iOS and Android rather than just returning JSON or HTML.
  • Streaming Interfaces: Screens are generated and rendered live as tokens stream from the LLM.
  • Platform-Aware Design: Automatically switches between Cupertino and Material 3 design systems based on the device.
  • Speculative Prefetching: Prefetches likely next destinations to make navigation between generated screens feel instant.
  • Tool Integration: Supports mid-stream tool calls (e.g., web search) to ground generated content in real-time data.

Related

  • Project
    AGenUI/AGenUI

    A high-performance A2UI SDK for building native generative UI experiences on iOS, Android, and HarmonyOS, enabling LLMs to render interactive components in real-time.

  • Project
    thesysdev/openui

    A full-stack Generative UI framework that uses a compact, streaming-first language to let LLMs render structured, interactive components with high token efficiency.

  • Project
    wandb/openui

    An open-source tool that lets you describe UI components in plain language and see them rendered live, with the ability to convert them to React, Svelte, or Web Components.

  • Project
    vercel-labs/json-render

    A Generative UI framework that enables AI to create safe, predictable, and cross-platform interfaces by constraining AI-generated JSON to a predefined component catalog.

  • Project
    droidrun/mobilerun

    An open-source framework that enables LLM agents to control Android and iOS devices using natural language, providing tools for UI inspection and mobile automation.