Skip to content
Free public beta until December 4, 2026: every course is open, no card needed. After that, the first chapter of each course stays free and full access is a one-time $39 ($19 for up to 100 founding members chosen for helpful feedback). Your progress is kept. How the beta works

LiteRT

Google's converter and file format for small, fast models; called TensorFlow Lite until 2024.

LiteRT (formerly TensorFlow Lite) converts trained models into a compact FlatBuffer (.tflite) of operators and weights, optionally quantized to int8, for phones and microcontrollers.

Examples

  • The TensorFlow Lite converter, renamed LiteRT in 2024, turns the trained gesture model into one compact .tflite FlatBuffer holding its operations and weights.

  • Given a representative sample of inputs, the same converter rewrites every number in the gesture model as an 8-bit integer.

See also

TensorFlow Lite Micro TinyML training
Read the full section →