https://mythic.ai/products/m1076-analog-matrix-processor/

Skip to content
 

  * Products
  * Technology
      + Compute-in-Memory
      + Dataflow Architecture
      + Analog Computing
      + Mythic AI Workflow
  * Markets
      + Smarter cities and spaces
      + Intelligent machines and robots
      + Smart Home
      + AR/VR and Metaverse
      + Drones and aerospace
  * Blog
  * Company

Products >> M1076 Analog Matrix Processor

M1076 Analog Matrix Processor

Overview

The M1076 Mythic AMP(tm) delivers up to 25 TOPS in a single chip for
high-end edge AI applications. The M1076 integrates 76 AMP tiles to
store up to 80M weight parameters and execute matrix multiplication
operations without any external memory. This allows the M1076 to
deliver the AI compute performance of a desktop GPU while consuming
up to 1/10th the power - all in a single chip.

M1076

Features

Array of 76 AMP tiles, each with a Mythic Analog Compute Engine
(Mythic ACE(tm))

On-chip DNN model execution and weight parameter storage with no
external DRAM

Execution of models at higher resolution and lower latency for better
results

4-lane PCIe 2.1 interface with up to 2GB/s of bandwidth for
inferencing processing

19mm x 15.5mm BGA package

Capacity for up to 80M weights - able to run single or multiple
complex DNNs entirely on-chip

Deterministic execution of AI models for predictable performance and
power

Support for INT4, INT8, and INT16 operations

Available I/Os - 10 GPIOs, QSPI, I2C, and UARTs

Typical power consumption running complex models 3~4W

Workflow

DNN models developed in standard frameworks such as Pytorch, Caffe,
and TensorFlow are implemented and deployed on the Mythic Analog
Matrix Processor (Mythic AMP^TM) using Mythic's AI software workflow.
Models are optimized, quantized from FP32 to INT8, and then retrained
for the Mythic Analog Compute Engine (Mythic ACE^TM) prior to being
processed through Mythic's powerful graph compiler. Resultant
binaries and model weights are then programmed into the Mythic AMP
for inference. Pre-qualified models are also available for developers
to quickly evaluate the Mythic AMP solution.

Mythic AI Workflow

DNN Model Library

Mythic provides a library of pre-qualified DNN models for the most
popular AI use cases. The DNN models have been optimized to take
advantage of the high-performance and low-power capabilities of the
Mythic Analog Matrix Processor (Mythic AMP^TM). Developers can focus
on model performance and end-application integration instead of the
time-consuming model development and training process. Available
pre-qualified DNN models include:

[Object_detector_board_1-300x195]
Object Detection and Classification YOLOv3, YOLOv5, ResNet-50,
ResNet-18
[downtown_detector-263x300]
Object Detection and Classification YOLOv3, YOLOv5, ResNet-50,
ResNet-18
[Room-Segmentation_v2-300x238]
Scene Segmentation SegNet
[Human-pose-estimator-300x200]
Human Pose Estimator OpenPose Body25

Resources

 
[Gold5-1024x559]

Mythic Whitepaper / Blog

Download
 
[m1076-brief]

M1076 Mythic AMP(tm) Product Brief

Download
 
[Gold5-1024x559]

Mythic Whitepaper / Blog

Download
Return to Portfolio
Product Inquiry
Mythic Logo

Mythic is enabling power-efficient AI acceleration for the edge

ANAB Accredited

Social

Twitter Facebook Linkedin Glassdoor

(c) 2023 Mythic (r). All rights reserved | Privacy Policy