DX-M1M
Real-time, on-device intelligence will be
universal, affordable, and transformative
Overview
Experience the future of on-device AI with the DEEPX DX-M1M. This AI processor delivers a stunning 25 TOPS while maintaining an ultra-low power consumption of just 3W(Typical). We prioritize Inferences Per Second (IPS) per Watt, so every operation delivers maximum, real-world value.
The DX-M1M integrates effortlessly with your host system, accelerating diverse applications like computer vision and audio analysis.
We believe that powerful AI should be accessible to all. Simply use your models from popular frameworks like Ultralytics Yolo 26, TensorFlow, PyTorch or ONNX, and our tools will automatically optimize them for the DX-M1M. This allows you to focus on innovation, not integration.
Features | Details |
|---|---|
AI Performance | 25 TOPS (INT8) |
Host Interface | PCIe Gen3 x2 |
Supports Gen 1/2/3 & x1/x2 | |
Memory | Integrated 1GB LPDDR4x (4266 MT/s) |
Power Consumption | 3W (Typical) |
Operating Temperature | -25 ~ 85°C (Throttling), -25 ~ 65°C (Non-Throttling) |
Form Factor | M.2 2242 (M+B Key), 22 x 42mm |
Software Support | Windows 11, 10 64 bit |
Ubuntu 22.04, 20.04 LTS
Support Yocto Project and Docker | |
Support TensorFlow, ONNX, Keras, PyTorch by Dataflow complier converted | |
System Support | x86, ARM Based Architecture |
Key Benefits
Unprecedented Power EfficiencyThe DX-M1M is a groundbreaking AI processor that redefines on-device intelligence. It delivers a powerful 25 TOPS in a 3W ultra-low power design, prioritizing Inferences Per Second (IPS) per Watt for maximum, real-world value. It's the perfect solution for power-sensitive edge devices.
A Seamless Developer ExperienceWe believe that powerful AI should be easy to use. The DX-M1M comes with a comprehensive software environment. This tool automatically optimizes models from popular frameworks like PyTorch and ONNX, so you can focus on innovation, not complex integration.
Innovative, AI-Optimized ArchitectureThe DX-M1M features a dedicated NPU (Neural Processing Unit), an architecture built specifically to accelerate deep learning. Its specialized hardware for common AI tasks like pooling and Softmax enables faster, more efficient processing and true real-time on-device AI.
Robust, Hardware-Based SecurityProtecting your AI is our priority. DX-M1M’s robust, hardware-based security system features a Root of Trust and a True Random Number Generator (TRNG), ensuring your AI models and data are protected from a wide range of external threats.
Related Products
DX-M Series
AI Chips
The core NPU delivering high performance and accuracy with industry-leading power efficiency
DX-M Series
AI Modules
Ultra-compact M.2 modules that simplify edge AI deployment with integrated NPU and memory
DX-M Series
Industrial PCs
Use Cases
AI CCTV
Enables real-time object detection and behavioral analysis directly on the camera.
ADAS
Accelerates real-time recognition of traffic signs, pedestrians, and lanes for enhanced safety.
Industrial Robot Arms
Performs real-time visual inspection and precise assembly for quality control in manufacturing.
AI Dashcams
Provides advanced incident detection and video analysis to assist drivers.
Autonomous Mobile Robots (AMRs)
Handles navigation, object avoidance, and logistics tasks in warehouses and delivery services.
Drones
Provides autonomous navigation and aerial surveillance by processing video feeds on-device.
Edge Servers
Manages and processes data from multiple sensors and cameras locally, reducing latency and reliance on the cloud
AIoT Devices
Powers smart home and retail devices with on-device data analysis and real-time intelligence.
Remote Patient Monitoring (RPM)
Real-time, on-device analysis of patient vital signs to immediately detect and alert for emergencies like abnormal heart rhythms or falls.