M5Stack LLM (Large Language Model) Module (AX630C) [Discontinued]

This LLM module is trained on huge amounts of data (hence the name large) making it able to understand and generate natural language and other types of content to perform a wide range of tasks, such as wake word detection (Detects phrases/keywords in audio streams and conversations), speech recognition (ASR), multilingual support for over 29 languages and text-to-speech (TTS). With the StackFlow framework and Arduino/UiFlow libraries on board, It lets you easily carry out smart features with minimal coding effort. With an AX630C SoC processor, the module also includes a 3.2 TOPs (Tera Operations per Second) NPU with native support for Transformer models that measure and simplify the AI's performance. It has 4GB LPDDR4 memory (1GB for what you want to use it for and 3GB for hardware acceleration) and 32GB eMMC storage, supporting the parallel loading and sequential inference of multiple models for straightforward multitasking. Here are some of the handy bits that come on board. A microphone, speaker, SD card slot, USB OTG, and an RGB status light allow you to record, listen back, upgrade the firmware and transfer data. There is also a UART communication port for easy connection and debugging and a USB-C port with master-slave auto-switching functionality for connecting additional devices such as cameras. Features

* Offline inference, 3.2T@INT8 precision computing power

* Integrated KWS (wake word), ASR (speech recognition), LLM (large language model), TTS (text-to-speech generation)

* Multi-model parallel processing

* 32GB eMMC storage and 4GB LPDDR4 memory

* Microphone and speaker

* Serial communication

* SD card firmware upgrade

* Supports ADB debugging

* RGB indicator light

* Built-in Ubuntu system

* Supports OTG functionality

* Compatible with Arduino/UIFlow

Specifications

Processor SoC AX630C@Dual Cortex A53 1.2 GHz

MAX.12.8 TOPS @INT4 and 3.2 TOPS @INT8

Memory 4GB LPDDR4 (1GB system memory + 3GB dedicated for hardware acceleration)

Storage 32GB eMMC5.1

Communication Serial communication default baud rate 115200@8N1 (adjustable)

Microphone MSM421A

Audio Driver AW8737

Speaker 8Ω@1W, Size:2014 cavity speaker

Built-in Units KWS (wake word), ASR (speech recognition), LLM (large language model), TTS (text-to-speech)

RGB Light 3x RGB LED@2020 driven by LP5562 (status indication)

Power Idle: 5V@0.5W, Full load: 5V@1.5W

Button For entering download mode for firmware upgrade

Upgrade Port SD card / Type-C port

Working Temp 0-40°C

Product Size 54 x 54 x 13mm

Packaging Size 133 x 95 x 16mm

Product Weight 17.4g

Packaging Weight 32.0g

Resources

* Product Wiki

* Arduino IDE Tutorial

* UIFlow2.0 Tutorial

* Schematic

Pinout

Package Contents

* 1x LLM Module