Nature Quest (pocket edition): Bring Mindfulness Outdoor.
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass . Nature Quest (pocket edition) is taking Gemma 4 E2B outside into the outdoor environment with no internet connection, fully Offline. What I Built Pocket Nature Quest is a fully offline Android field companion that turns one outdoor photo into a short Look · Listen · Move quest. The idea is deliberately a…
Nature Quest (pocket edition) is an Android field companion that operates on-device without the need for an internet connection, API keys, or online geocoding. It aims to promote mindfulness and wellness by encouraging outdoor exploration through a simple photo-based quest system. The app employs the locally-run Gemma 4 E2B model to generate a short, three-part action plan based on the user's chosen photo, time, and energy level.
The user takes a photo, optionally adds GPS context, selects an appropriate time and energy level, and receives a short list of actions to complete outdoors. The app was designed for a variety of users, including walkers, families, travelers, and nature beginners, who want to enhance their observation skills of their surroundings.
The app does not require a Wi-Fi connection, API keys, web search, or online geocoding to function. The project initially began with a misunderstanding in the initial prompt, leading to a fast web prototype and the final Android application. The Android application, however, is the version that was submitted for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass.
The app was tested at Sunset Pier, a waterfront area with space to walk, take photos, and sit, dine, and drink coffee. The results showed that the local Gemma 4 E2B model produced more specific and articulated output compared to the cloud model, while still maintaining the essential features of looking at a real photo, creating a safe three-part activity, and ensuring data ownership and offline availability.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.