TrailMate AI: An Open-Source AI Companion That Gets You Outside

Description and use

TrailMate AI is an open‑source app that uses a local LLM to generate short outdoor activity plans, encouraging users to step outside and disconnect from screens.

Problem

People often lack ideas or motivation to spend time outdoors, leading to screen‑bound habits.

Outcome

TrailMate AI demonstrates how a locally hosted LLM can generate actionable outdoor plans, reducing screen time and encouraging real‑world exploration.

Workflow

• 1. User selects activity (e.g., nature walk, birdwatching).
• 2. User inputs available time and location description.
• 3. Frontend sends request to Node/Express backend.
• 4. Backend forwards prompt to local Ollama service using qwen2.5:3b model.
• 5. Model returns a short activity plan with practical steps and prompts.
• 6. Frontend displays the plan to the user.

Evidence

I built TrailMate AI, a small outdoor companion designed to turn a little curiosity into time spent outside.

Source

Users can choose an activity such as a nature walk, birdwatching, gardening, or a mindful outdoor break.

Source

The app is designed as a small, self-hostable project that developers can modify to use another locally available model.

Source

For AI inference, it connects to Ollama and uses the open-weight qwen2.5:3b model by default.

Source