FitQuest AI: Turning Screen Time into Outdoor Adventures with Local Open-Source AI

Description and use

FitQuest AI leverages a local qwen2.5:3b model to generate outdoor quests, encouraging users to reduce screen time while preserving privacy.

Problem

Developers and desk workers spend too much time on screens and need motivation to go outside for physical activity.

Outcome

FitQuest AI demonstrates that local open‑source AI can motivate users to reduce screen time by turning outdoor activity into gamified quests, while preserving privacy and lowering inference costs.

Workflow

• 1. User installs FitQuest AI locally. 2. App downloads the qwen2.5:3b model via Ollama. 3. Backend FastAPI connects to Ollama and validates AI output with Pydantic. 4. AI generates a structured quest based on user goals and recent activity. 5. User goes outdoors, completes the quest, and logs the workout. 6. Backend calculates XP, updates streaks, and unlocks achievements. 7. User can capture trail photos and generate shareable story cards via an HTML Canvas renderer.

Evidence

FitQuest AI uses local AI to turn outdoor movement into bite‑sized quests—then encourages you to put your screen away and enjoy the real world.

Source

Local inference helps keep workout context away from third‑party cloud AI inference services.

Source

The backend connects to an open‑weight model (qwen2.5:3b) through Ollama. Pydantic validates the generated structured data before it is used by the application.

Source