system-on-module

Grinn GenioSOM-510

The Grinn GenioSOM-510 is an advanced System on Module (SoM) designed to meet the performance and efficiency needs of Artificial Intelligence and Internet of Things applications. Leveraging a powerful MediaTek® Genio 510 (MT8370) IoT application processor, this SoM is ideal for edge computing scenarios that demand real-time processing and multimedia capabilities.

Wiki page

Datasheet

Fields of application

Industrial Applications
Smart Home
Multimedia systems
Image Recognition

Supporting

Specification

Processor
- CPU: 2x Arm Cortex-A78, 4x Arm Cortex-A55
- GPU: Arm Mali-G57 MC2
- NPU: MediaTek 5th Gen., 3.2 TOPS
- AudioDSP: Cadence Tensilica HiFi 5
Memory and Storage
- 2x 2GB LPDDR4 RAM (optional 2x 4GB)
- 16GB eMMC (up to 64GB)
Video Input
- 2x MIPI CSI-2 (4L)
- Virtual channel: 8x FHD30
Display
- 1x HDMI 2.0
- 1x DP 1.4
- 1x eDP 1.2
- 2x MIPI-DSI (4L)
- Dual display, max. 4K60 + FHD60
Interfaces
- 2x USB 2.0, 1x USB 3.1 Gen 1
- 1x Eth Gb MAC
- 1x PCIe 2.0 (1L)
- 4x UART
- 5x I2C, 2x I2C/I3C
- 6x SPI
Audio
- 2x I2S
- Support for external audio codec integration
Power
- Input voltage 3.15V .. 5.0V
Form Factor
- Dimensions: 37 x 42.6 mm
- LGA312 package
Ambient operating temperature
- -20°C to 70°C (Extended)
- -40°C to 85°C (Industrial)
Multimedia
- Video playback: 4K@60fps H.264/H.265
- Video encoding: 1080p@60fps H.264/H.265
Connectivity
- Gigabit Ethernet (via external PHY)
- Wi-Fi / BT support via external module
Video Decode
- H.264 / H.265 / VP9 / AV1 (4K60)
- MPEG4 / VP8 (FHD60)
Video Encode
- H.264 / H265 (4K30)

ready-to-use

Other products

Grinn GenioBoard + DEEPX

Delivering 25 TOPS of native INT8 performance, the Grinn GenioBoard with the DEEPX DX-M1 M.2 Module enables high-throughput AI inference directly at the edge.

Grinn AstraOneSBC-1680 + DEEPX

Achieving 25 TOPS of native INT8 compute performance with the DEEPX DX-M1 M.2 accelerator alongside the integrated 8 TOPS NPU, the AstraOneSBC-1680 platform is built for compute-intensive edge AI workloads that demand higher inference capacity.

Grinn AstraSOM-1680

System on Module featuring a quad-core Cortex A73 processor, advanced GPU, and a neural network accelerator. Ideal for high-performance and AI-driven applications.