This repository features a generative AI image generator designed to run on the RP2350 microcontroller (Raspberry Pi Pico 2). The model uses a latent flow diffusion transformer (DiT) architecture, similar to Flux, but scaled down significantly with only 1.7 or 2.9 million parameters—thousands of times smaller than typical local diffusion models. It can generate $128 times 128$ RGB images of human faces in approximately 5-20 seconds and supports conditional generation for five classes based on gender, smile, and neutral expressions.
- The model architecture includes a VAE decoder and a DiT component.
- Hardware output can be displayed via a VGA monitor or streamed over USB.
- Includes an int8 pipeline involving calibration, distillation-QAT (Quantization-Aware Training), and folding.
- Two variants are provided: a high-quality flagship model and a faster build version.
Paris-based Open Tools is developing an open-source inkjet printer that uses standard mechanical components and refillable ink cartridges, controlled by a Raspberry Pi Zero W. It can print on both sheets and rolls of paper, automatically cutting the latter to size, and aims to provide a repairable, upgradeable alternative to traditional printers.
This document provides a developer guide for the Tiny Code Reader from Useful Sensors, a small, low-cost hardware module that reads QR codes. It covers connecting, mounting, powering up, reading data, configuration, sensor characteristics, example code, privacy considerations, and an appendix with data formats and CAD files.
Open-source project showcasing a speech-to-text application on a low-cost CH32V003 microcontroller. The application reads audio from an analog microphone, identifies spoken digits 'zero' to 'nine', and logs the results over a serial link.