tinyml

Machine Learning Systems: Foundations, Scaling, Agentic AI, and Physical AI (Vols I–IV) • Harvard CS249r | https://mlsysbook.ai

Automation foundation model for tiny devices: 2-bit, 8-29 MB, tool calls, structured extraction and embeddings on phones, wearables, smart homes, robots, cars and microcontrollers.

[ICLR 2020] Once for All: Train One Network and Specialize it for Efficient Deployment

Z80-μLM is a 2-bit quantized language model small enough to run on an 8-bit Z80 processor. Train conversational models in Python, export them as CP/M .COM binaries, and chat with your vintage computer.

Machine Learning inference engine for Microcontrollers and Embedded devices

[NeurIPS 2020] MCUNet: Tiny Deep Learning on IoT Devices; [NeurIPS 2021] MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learning

Seeed SenseCraft Model Assistant is an open-source project focused on embedded AI. 🔥🔥🔥

A ternary, zero-heap tiny language model that runs inside a $2 microcontroller — bit-exact Python <-> C99 <-> Cortex-M3 (QEMU) parity. Apache-2.0.

Building Simple versions of AI (ML, DL, NN) models from scratch to help grasp the concepts

A £20 open-source rotary inverted pendulum that balances itself with a 689-parameter neural network running standalone on an Arduino Nano.

DNN Compiler for Heterogeneous SoCs

SensiML's open-source AutoML solution for Edge AI model development

559 K パラメータの日本語 TTS を ESP32-S3 で実時間合成。漢字かな交じり文の形態素解析・アクセント推定まで端末内で走る（M5Stack CoreS3 実機で確認）。推論は依存ゼロの C99、ブラウザ demo あり。arXiv:2608.21378 sanoTTS の日本語 clean-room 再実装。⚠️ コードは MIT ですが、配布モデルの重みは MIT ではありません（LICENSE-MODEL.md。出力に用途制限が伝播します）

Ultra-tiny 2.1k streaming voice activity detector (2.1 KB INT8). 200 ms window for microcontrollers, edge devices, and DSPs. ships with ONNX, TorchScript, and standalone C header.

TinyML & Edge AI: On-device inference, model quantization, embedded ML, ultra-low-power AI for microcontrollers and IoT devices.

Run a 28.9M-parameter TinyLM on ESP32-S3 with an RP2040 OLED display node for fully local embedded AI inference.

Source code generators for machine learning models.

[Tiny VAD] SG-VAD: Stochastic Gates Based Speech Activity Detection

Code for WF-IoT paper 'TinyML Benchmark: Executing Fully Connected Neural Networks on Commodity Microcontrollers'

Jamming Detection in IoT Wireless Networks using TinyML -- Code and Dataset