Home > Rockchip Products > AI Modules > 
DeepX DX-M1 M.2 AI Module
200 GPU-class TOPS
USD Bulk Pricing
Fast Shipping
DX-M1 AI Chip 25 TOPS 4GB LPDDR5
∞
Sold
7x24 Hours
Technical Support
1 V 1
Service
Product Highlights:
- Performance：25 TOPS (INT8)
- Interface：PCIe Gen3 x4 (Supports Gen 1/2/3 & x1/x2/x4)
- Memory: 4GB LPDDR5 (5600 MT/s)
- Power Consumption: 1W (min) ~ 5W (max)
- Operating Temperature: -40 ~ 85°C
- Form Factor: M.2 2280(M Key), 22 x 80mm
- OS Support: Windows 10/11, Linux (Ubuntu 24.04/22.04/20.04 LTS),
Quote
Matching package
USD Bulk Pricing
Fast Shipping
DX-M1 M.2 Module
Key features
- AI Performance with 25 TOPS (INT8)
- PCIe Gen3 x4 (Supports Gen 1/2/3 & x1/x2/x4)
- 4GB LPDDR5 (5600 MT/s) Memory
- M.2 2280(M Key), 22 x 80mm
The DX-M1 M.2 AI Module embodies advanced silicon technology, demonstrating that intelligence can be deployed at the edge, independent of cloud infrastructure. Operating within a power envelope of just 1–5W, it delivers a processing throughput of approximately 200 GPU-class TOPS, enabling edge devices to transition from passive data collectors to autonomous decision-makers through an innovative energy-efficient architecture. Its proprietary memory organization supports concurrent execution of multiple AI models, establishing a new baseline for embedded AI platforms. This integration reduces reliance on costly, high-power computational systems, making sophisticated AI capabilities accessible across a diverse range of devices, industrial environments, and urban infrastructures.
We authored an article discussing the specifications and features of the DX-M1 M.2 and DX-M1M M.2.
Target Applications
- Edge Cameras Systems
- Smart Mobility
- Smart Factory
- Smart Cities
- Robotics
- Drones
- Edge Computing
- Smart Homes
- Smart Retail
