Buy 5 or More Units - Unlock Priority Discounts!
Buy 20 or More Units - Get Exclusive Volume Discounts!
Also Support:
Are you a reseller or placing a large order?
We're happy to offer Special Pricing tailored to your needs.
Click "Request Quote" or email us at [email protected]
to get your best price today!
Product Code: DBHO-124
MPN #: G650-07968-01
The Coral G650-07968-01 Dev Board Micro is a compact development board designed for embedded systems, machine learning, and edge AI applications. Powered by the NXP i.MX RT1176 microcontroller and a Coral Edge TPU coprocessor, it enables developers to build intelligent devices that process data locally without relying entirely on cloud-based computing.
Featuring dual Arm Cortex-M7 and Cortex-M4 cores, the board supports a range of embedded computing tasks, from real-time control to sensor data processing. Its support for TensorFlow Lite for Microcontrollers makes it suitable for developing machine learning applications in resource-constrained environments, including smart sensors, connected devices, and intelligent control systems.
The integrated Coral Edge TPU accelerator delivers up to 4 trillion operations per second (TOPS) for high-performance inference with TensorFlow Lite models. This capability helps developers implement applications such as image recognition, object detection, and other supported machine learning tasks directly on the device, enabling responsive processing with reduced dependence on external computing resources.
With an onboard Himax HM01B0 camera sensor, a mono microphone, and multiple hardware connection options, the Dev Board Micro supports prototyping for vision-based and audio-enabled applications. Its GPIO headers and board-to-board connectors provide interfaces for connecting peripherals, sensors, and compatible expansion boards, making it useful for robotics, automation, and embedded product development.
The compact design, onboard memory, secure element, and USB-C connectivity make the Coral G650-07968-01 suitable for developers, engineers, and researchers building space-constrained intelligent devices. It provides a practical platform for experimenting with embedded machine learning, sensor integration, and edge inference in applications where size and local processing capabilities are important.
Coral Edge TPU accelerator delivering up to 4 TOPS for machine learning inference
Supports TensorFlow Lite for Microcontrollers
2 MB SRAM, 64 MB SDRAM, and 128 MiB NAND flash memory
Integrated Himax HM01B0 camera sensor and mono PDM microphone
NXP A71CH secure element for device authentication and secure credential storage
GPIO headers and dual 100-pin board-to-board connectors
Compact 65.0 × 30.0 × 6.8 mm design weighing 10.4 g
| NXP i.MX RT1176 MCU | |
| ARM Cortex-M7 | ● 800 MHz ● 32 KB L1 instruction cache ● 32 KB L1 data cache ● Floating Point Unit (FPU) with single-precision and double-precision support of Armv7-M architecture FPv5 ● Supports TensorFlow Lite for Microcontrollers |
| ARM Cortex-M4 | ● 400 MHz ● 16 KB instruction cache ● 16 KB data cache ● Single-precision FPU defined by Armv7-M architecture FPv4-SP ● Supports TensorFlow Lite for Microcontrollers |
| Memory | ● 2 MB SRAM: 512 KB of TCM for Cortex-M7 256 KB of TCM for Cortex-M4 1.25 MB OCRAM ● 4 KB secure always-on RAM ● 256 KB boot ROM |
| Secure element | |
| NXP A71CH | Cryptographic device authentication |
| NXP A71CH | Secure credentials provisioning and storage |
| ML accelerator | |
| Coral Edge TPU coprocessor | High performance inferencing for TensorFlow Lite models |
| Coral Edge TPU coprocessor | 4 trillion operations per second (TOPS) |
| Coral Edge TPU coprocessor | USB 2.0 interface with the RT1176 MCU |
| On-board sensors | |
| Camera | ● Himax HM01B0 CMOS sensor ● 324 x 324 px native resolution, or QVGA (320 x 240) ● 110º diagonal FOV ● ƒ/2.0 focal ratio ● Fixed focus |
| Microphone | Mono microphone (PDM) |
| Memory and storage | |
| Random access memory (SDRAM) | 64 MB |
| Flash memory (NAND) | 128 MiB |
| Hardware connections | |
| GPIO headers | 2x 12-pin through-hole interface |
| GPIO headers | GPIO, PWM, UART, I2C, SPI, DAC, ADC, GND, 3V3, 1V8, and VSYS |
| Board-to-board connectors | 2x 100-pin connectors (Hirose DF40C-100DP-0.4V(51)) Provides various I/O from the MCU |
| USB-C plug | Board power and USB2 data |
| User interface | |
| LEDs | 4 on-board LEDs: Green LED that indicates camera activity Green LED that's user programmable Orange LED that indicates board operating status White LED that indicates Edge TPU operating status |
| Switches | 2 switch buttons |
| Measurement | |
| Size | 65.0 × 30.0 × 6.8 mm |
| Weight | 10.4 g |
Be the first to write a review
We are retiring PICOPC Products MNHO-073, MNHO-048 and MNHO-043 and they have reached EOL (End Of Life) as we are introducing new products with better technology. We would recommend you to look into MNHO-095. For Untangle, pfSense, OPNsense, Windows 11, Proxmox VE etc. we recommend MNHO-095 whereas for Sophos, Kerio Control and Endian, we would recommend MNHO-069. (A complete list of supported OS can be found in each product's listing)