Overview
This guide documents experiments with using Large Language Models (LLMs) to summarize and provide interesting insights into data collected from an environmental sensor. It is not a fully fleshed out project that is ready to automate all of your environmental needs. It is still early days in the world of local and small LLMs. The goal with this guide is to evaluate different models and methods to find what current possibilities are, and see what kinds of things work and don't work.
The experiments are fully open source. Code used to collect data, format it, and the prompts used are all made available. Do you know a model that is better for this type of task, or have ideas for better ways to prompt or feed data into the models? Feel free to reach out and share your own findings.
These experiments were conducted with the SEN6x environmental sensor, and accompanying Adafruit SEN6x Breakout. The general principles of this experiment could of course be re-created or expanded upon with any other sensors that you have handy.
The SEN6X environmental sensor is not stocked by Adafruit. You can buy it from authorized distributors. Below is a link to the sensor available at DigiKey. You'll want the sensor (SEN66), not the development kit.
Other parts from Adafruit:
Page last edited September 23, 2025
Text editor powered by tinymce.
