FUTO Swipe

Open models for on-device swipe typing

Custom KeyboardsOpen SourceUser Experience

▲ 0 votes1 commentsLaunched Jun 24, 2026

Visit Website

Daily #2Weekly #43

FUTO Swipe is a family of small, open models for accurate swipe typing. It includes a layout-agnostic encoder, a layout-specific decoder, and a lightweight context language model. The full system runs efficiently on-device with a very small footprint, and FUTO has also released the 1 million swipe dataset used to train it.

AI Analysis

📝 Summary

FUTO Swipe offers a family of small, open-source models for accurate on-device swipe typing. It includes a layout-agnostic encoder, layout-specific decoder, and lightweight context language model that runs efficiently with a minimal footprint. It solves key pain points like privacy risks from cloud-based keyboards (e.g. data collection by Gboard) and lack of high-quality gesture typing in open-source alternatives. By releasing the models and the 1 million swipe dataset, it promotes transparency, customization, and community innovation. The value proposition is enabling private, efficient, and accessible advanced typing for developers building custom keyboards and privacy-focused users.

📈 Market Timing

The market timing is favorable for 2025-2026 due to maturing on-device AI technologies, rising privacy regulations and user concerns over data collection, growing demand for open-source alternatives, and the trend toward efficient small models amid AI democratization. This aligns well with increasing interest in customizable, private mobile UX. Excellent Timing.

✅ Feasibility

High. Models are small, proven to run efficiently on-device with low compute needs, keeping development and operation costs minimal. Open-source approach reduces risks; compliance is straightforward as no cloud data handling. High scalability for integration into keyboards, though adoption depends on developer uptake and cross-layout accuracy.

🎯 Target Market

Main target segments: open-source developers and creators of custom Android keyboards, privacy-conscious tech enthusiasts and power users (ages 18-45). Industries: mobile input methods, AI tools, open-source software. Geographic: global with strong adoption potential in US and Europe due to privacy awareness. Core pain points: insufficient swipe accuracy in non-proprietary keyboards and data privacy risks. Willingness to pay: moderate, via project donations, premium apps or integration services.

⚔️ Competition

Low. Direct competitors: 1. Gboard (gboard.app), 2. Microsoft SwiftKey (swiftkey.com), 3. FlorisBoard (florisboard.org), 4. AnySoftKeyboard (anysoftkeyboard.github.io), 5. TouchPal (touchpal.com). Advantages: fully open-source with released dataset, guaranteed on-device privacy, small efficient footprint. Disadvantages: may require more developer effort to integrate, potentially less polished accuracy or language support than proprietary cloud solutions initially.

Upgrade Pro to unlock full AI analysis