Summary of MAXIM INTEGRATED MAX78000 AI MICROCONTROLLER
The MAX78000 is an ultra-low-power system-on-chip designed for edge AI, combining an Arm Cortex-M4 microcontroller with a hardware-based CNN accelerator. It features 512KB Flash and 128KB SRAM, supporting battery-powered applications with microjoule energy consumption. The device includes a RISC-V coprocessor, extensive I/O pins, and interfaces like I2S and PCIF, available in compact BGA and WLP packages to enable efficient neural network execution at the IoT edge.
Parts used in the MAX78000 Project:
- Arm Cortex-M4 Processor with FPU
- 512KB Flash Memory
- 128KB SRAM
- 16KB Instruction Cache
- 32-Bit RISC-V Coprocessor
- CNN Accelerator Engine
- I2S Interface
- Parallel Camera Interface (PCIF)
- General-Purpose I/O Pins
MAX78000 – Ultra-low-power Arm Cortex-M4 processor with FPU-based microcontroller with Convolutional Neural Network Accelerator.
The MAX78000 is is an advanced system-on-chip built to enable neural networks to execute at ultra-low power and live at the edge of the IoT. This product combines the most energy-efficient AI processing with Maxim’s proven ultra-low power microcontrollers. The hardware-based convolutional neural network (CNN) accelerator enables battery-powered applications to execute AI inferences while spending only microjoules of energy.
In addition to the memory in the CNN engine, the MAX78000 has large on-chip system memory for the microcontroller core, with 512KB flash and up to 128KB SRAM. Multiple high-speed and low-power communications interfaces are supported, including I2S and a parallel camera interface (PCIF).
The device is available in 81-pin CTBGA (8mm x 8mm, 0.8mm pitch) and 130-pin WLP (4.6mm x 3.7mm, 0.35mm pitch) packages.
Key Features
Dual Core Ultra-Low-Power Microcontroller
- Arm Cortex-M4 Processor with FPU Up to 100MHz
- 512KB Flash and 128KB SRAM
- Optimized Performance with 16KB Instruction Cache
- Optional Error Correction Code (ECC-SEC-DED) for SRAM
- 32-Bit RISC-V Coprocessor up to 60MHz
- Up to 52 General-Purpose I/O Pins
- 12-Bit Parallel Camera Interface
- One I2S Master/Slave for Digital Audio Interface
Neural Network Accelerator
- Highly Optimized for Deep Convolutional Neural Networks
- 442k 8bit Weight Capacity with 1,2,4,8-bit Weights
- Programmable Input Image Size up to 1024 x 1024 pixels
- Programmable Network Depth up to 64 Layers
- Programmable per Layer Network Channel Widths up to 1024 Channels
- 1 and 2 Dimensional
Read more: MAXIM INTEGRATED MAX78000 AI MICROCONTROLLER
- What is the primary function of the MAX78000?
The MAX78000 enables neural networks to execute at ultra-low power and live at the edge of the IoT. - How much memory does the MAX78000 have?
The device has 512KB flash and up to 128KB SRAM. - Does the MAX78000 support digital audio interfaces?
Yes, it supports one I2S Master/Slave for a Digital Audio Interface. - Can the MAX78000 run AI inferences on battery power?
Yes, the hardware-based CNN accelerator allows battery-powered applications to execute AI inferences while spending only microjoules of energy. - What is the maximum input image size supported by the CNN engine?
The programmable input image size goes up to 1024 x 1024 pixels. - Is there a coprocessor included in the MAX78000?
Yes, it includes a 32-Bit RISC-V Coprocessor running up to 60MHz. - How many general-purpose I/O pins are available?
The device offers up to 52 General-Purpose I/O Pins. - What package options are available for the MAX78000?
The device is available in 81-pin CTBGA and 130-pin WLP packages. - Can the MAX78000 support error correction for SRAM?
Yes, it offers optional Error Correction Code (ECC-SEC-DED) for SRAM. - What is the maximum number of layers the network can support?
The programmable network depth supports up to 64 Layers.
