phyFLEX®-i.MX 93
Energy-efficient edge computing, accelerated Machine Learning
Product Overview
Features
Energy Efficient Edge Processing
On-Device Machine Learning
The integrated Arm Ethos™-U65 NPU provides dedicated acceleration for machine learning workloads, enabling local inference for applications such as classification, detection, and sensor processing. By processing AI workloads directly on the device, applications can reduce latency and dependence on cloud connectivity.
Application and Real-Time Processing
Secure and Flexible Embedded Processing
Specifications
Processor |
|
| Processor | NXP i.MX 93 |
| Architecture | Arm Cortex-A55 / Arm Cortex-M33 |
| Frequency | 1.7 GHz, 250 MHz |
| CPU | Arm NEON™ |
Memory |
|
| RAM | 1 GB / 2 GB max LPDDR4 |
| eMMC | 256 GB max |
| EEPROM | 4 KB / 32 KB max |
| SD/SDIO/MMC Interfaces | 2 |
High-Speed Interfaces |
|
| USB | 2x USB 2.0 OTG |
Peripherals |
|
| I3C | 1 |
| GPIO | 4, Up to 30 |
| I2C | 2, Up to 8 |
| PWM | 2, Up to 24 |
| CAN | 2x CAN FD |
| SPI/SSP | 2, Up to 8 |
| UART | 3, Up to 8 |
Connectivity |
|
| Ethernet | 1x 10/100 Mbps, 1x GbE (RMII) (optional RGMII with TSN Support) |
Multimedia |
|
| Audio | 1x SAI, Up to 3x SAI |
| Camera | 1x MIPI CSI-2 |
| Display | 1x LVDS, 1x MIPI-DSI |
Miscellaneous |
|
| JTAG | Yes |
| RTC | Yes |
| Security Accelerators | Secure Boot, SNVS, EdgeLock®, SRTC, Tamper Detection, TRDC, TrustZone® |
| Tamper Protection | Yes |
Mechanical |
|
| Dimensions | 37 mm x 40 mm |
| Connector | 1.27 mm pitch, FPSC-Gamma, Fused Tin Grid Array (FTGA) |
Operating Conditions |
|
| Operating Temperature | -40° to +85° C |
| Power (Typical) | TBD |
| Power (Max) | TBD |
| Supply Voltage | 5.0V |
Hardware Documentation
SOM |
|
|
Hardware Manual
|
|
|
Component Placement
|
|
|
Schematic Symbol and Footprint
|
Coming Soon! |
Carrier Board |
|
|
Component Placement
|
|
The i.MX93 & i.MX91 Processors
The NXP® i.MX 93 and i.MX 91 applications processors combine energy-efficient Arm® Cortex®-A55 processing with advanced security, connectivity, and low-power operation for intelligent edge devices. The i.MX 93 adds a second Cortex-A55, Cortex-M33 real-time core, and Arm Ethos™-U65 machine learning acceleration for applications requiring local AI and real-time processing.
Pin-compatible i.MX 93 and i.MX 91 options give developers flexibility to select the processing capabilities needed for their application while maintaining a consistent hardware platform.
3D Model
Sketch-fab model, Coming Soon
Resources |
|
|
3D Files
SOM
|
|
|
3D Files
Carrier Board
|
|
Software
Linux |
|
|
BSP-Yocto-NXP-i.MX93-PD26.1.0
2026-03-13
|
Yocto Walnascar 5.2.2 |
Related Case Studies
Apollo Video (Luminator) – niiu Video Monitoring System
Luminator Technology Group, a leading global provider of passenger information systems, acquired Apollo Video in 2017.
iNOEX – Warp Portable
A mobile terahertz based hand-held system specifically designed for medium and large-sized extruded pipes.
Ready to get started?
Let’s go! Learn more about PHYTEC SOMs


