DX-M1 AI Accelerator
Intelligence liberated — powerful, efficient, and universal at the edge
The DX-M1 AI chip is more than silicon, it is the declaration that intelligence belongs everywhere, not just in the cloud. At only 2–5W, it delivers the force of 200 GPU-class TOPS, transforming the edge from a passive receiver into an active decision-maker. at just 2-5W through breakthrough efficiency design.
Through its innovative memory design, DX-M1 allows multiple AI models to work side by side, not as a luxury but as a new standard. What once demanded costly, power-hungry systems is now within reach of every device, every factory, every city.
This is not a step forward in chips. It is a step forward in civilization, where real-time intelligence becomes universal, affordable, and inevitable.
Target Applications
Edge Camera Systems, Smart Mobility, Smart Factory, Smart Cities, Smart Retail, Smart Homes, Robotics, Drones, Edge Computing
Specification
| Features | Details |
| Processor | 25 TOPS (=200 eTOPS / INT8) |
| Signal Interface | PCIe GEN3 x4 / Band width: 4GB/s (*Compatible to PCIE x1) |
| Power | 2W min., 5W max. for DEEPX supported AI models |
| Operating Temperature | -25 ~ 85°C (Throttling), -25 ~ 65°C (Non_Throttling) |
| Software and Framework Support | Windows 11, 10 64 bit |
| Ubuntu 22.04, 20.04 LTS Support Yocto Project | |
| Support TensorFlow, ONNX, Keras, PyTorch by Dataflow complier converted | |
| System Support | x86, ARM Based Architecture |
Performance
Real-world AI inference FPS(Frames Per Second) benchmarks comparing DEEPX with leading competitors
- This graph shows the benchmark results for Batch=1, all tested under the same conditions.
- Xavier measurements were conducted at room temperature on an NVIDIA Jetson AGX Xavier development kit. The benchmarks utilized INT8 precision with the “no batch” (default=1) option, as recommended by NVIDIA.
- Company A results used the optimize option of “random-calib-set” with default values for the parser and compiler.
- DEEPX DX-M1 numbers are based on the SDK (September 2025), DX-Compiler v.2.0.0, and DX-RT v.3.0.0.
- Company A and DX-M1 were measured at room temperature on a single device through a PCIe interface on an x86 Host PC (Intel® Core™ i7-14700K CPU @ 5.6GHz).
Related Documents
DEEPX DX-M1 Chip Brief
DEEPX DX-M1 M.2 LPDDR5x2 Module Brief