-
Description
BeagleY-AI is a low-cost, open-source, and powerful 64-bit quad-core single-board computer, equipped with a GPU, DSP, and vision/deep learning accelerators, designed for developers and makers.
Users can take advantage of BeagleBoard.org's provided Debian Linux software images, which include a built-in development environment. This enables the seamless running of AI applications on a dedicated 4 TOPS co-processor, while simultaneously handling real-time I/O tasks with an 800 MHz microcontroller.
BeagleY-AI is designed to meet the needs of both professional developers and educational environments. It is affordable, easy to use, and open-source, removing barriers to innovation. Developers can explore in-depth lessons or push practical applications to their limits without restriction.
Specifications
| Processor | TI AM67 with quad-core 64-bit Arm Cortex-A53, GPU, DSP, and vision/deep learning accelerators |
| RAM | 4 GB LPDDR4 |
| Wi-Fi | BeagleBoard BM3301 module based on TI CC3301 (802.11ax Wi-Fi) |
| Bluetooth | Bluetooth Low Energy 5.4 (BLE) |
| USB | • 4x USB-A 3.0 supporting simultaneous 5 Gbps operation • 1x USB-C 2.0 supports USB 2.0 device |
| Ethernet | Gigabit Ethernet, with PoE+ support (requires separate PoE+ HAT) |
| Camera/Display | 1x 4-lane MIPI camera/display transceivers, 1x 4-lane MIPI camera |
| Display Output | 1x HDMI display, 1x OLDI display |
| Real-time Clock (RTC) | Supports an external button battery for power failure time retention. It is only populated on EVT samples. |
| Debug UART | 1x 3-pin debug UART |
| Power | 5 V/5 A DC power via USB-C, with Power Delivery support |
| Power Button | On/Off included |
| PCIe Interface | PCI-Express Gen3 x1 interface for fast peripherals (requires separate M.2 HAT or other adapter) |
| Expansion Connector | 40-pin header |
| Fan connector | 1x 4-pin fan connector, supports PWM speed control and speed measurement |
| Storage | microSD card slot, with support for high-speed SDR104 mode |
| Tag Connect | 1x JTAG, 1x Tag Connect for PMIC NVM Programming |
