Products Archive - Mythic https://mythic.ai Mon, 03 Jun 2024 21:54:24 +0000 en-US hourly 1 M1076 Analog Matrix Processor https://mythic.ai/products/m1076-analog-matrix-processor/ Wed, 08 Dec 2021 18:55:07 +0000 https://mythic.ai/?post_type=product&p=134 The M1076 Mythic AMP™ delivers up to 25 TOPS in a single chip for high-end edge AI applications.

The post M1076 Analog Matrix Processor appeared first on Mythic.

]]>

M1076 Analog Matrix Processor

Overview

The M1076 Mythic AMP™ delivers up to 25 TOPS in a single chip for high-end edge AI applications. The M1076 executes models at higher resolution and lower latency, integrating 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 ACE™

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

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 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 AMP™ using Mythic’s AI software workflow. Models are optimized, quantized from FP32 to INT8, and then retrained for the Mythic ACE™ 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 are optimized to take advantage of the high-performance and low-power capabilities of the Mythic AMP™. 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 Detection and Classification YOLOv3, YOLOv5, ResNet-50, ResNet-18
Object Detection and Classification YOLOv3, YOLOv5, ResNet-50, ResNet-18
Scene Segmentation SegNet
Human Pose Estimator OpenPose Body25

Resources

Mythic Whitepaper / Blog

Download

M1076 Mythic AMP™ Product Brief

Download

Mythic Whitepaper / Blog

Download
Return to Portfolio
Product Inquiry

The post M1076 Analog Matrix Processor appeared first on Mythic.

]]> MM1076 M.2 M Key Card https://mythic.ai/products/mm1076-m-2-m-key-card/ Wed, 08 Dec 2021 18:54:51 +0000 https://mythic.ai/?post_type=product&p=133 The MM1076 M.2 M key card enables high-performance, yet power-efficient AI inference for edge devices and edge servers.

The post MM1076 M.2 M Key Card appeared first on Mythic.

]]>

MM1076 M.2 M Key Card

Overview

The MM1076 M.2 M key card enables high-performance andpower-efficient AI inference for edge devices and edge servers. The M.2 card’s compact form-factor and popularity makes integration into many different systems a straightforward task. The MM1076 is designed with the M1076 Mythic AMP™ which is arranged in an array of AMP tiles that each feature a Mythic ACE™. The MM1076 is ideal for processing deep neural network (DNN) models in a variety of applications, including video surveillance, industrial machine vision, drone, AR/VR, and edge servers.

MM1076 M.2 M key card

Features

M1076 Mythic AMP™ with support for up to 80M weights on-chip

No external DRAM required

SMBus for EEPROM and PMIC access

Pre-qualified networks including object detectors, classifiers, pose estimators, with more being added

OS Support: Ubuntu, NVIDIA L4T, and Windows (future release)

Model parameters stored and matrix operations executed on-chip by AMP tiles

4-lane PCIe 2.1 for up to 2GB/s bandwidth

Support for standard frameworks, including PyTorch, TensorFlow 2.0, and Caffe

Small 22mm x 80mm form factor

Workflow

DNN models developed in standard frameworks such as Pytorch, Caffe, and TensorFlow are implemented and deployed on the Mythic Analog Matrix Processor (Mythic AMPTM) using Mythic’s AI software workflow. Models are optimized, quantized from FP32 to INT8, and then retrained for the Mythic Analog Compute Engine (Mythic ACETM) 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

Models

Mythic provides powerful pre-qualified models for the most popular AI use cases. Models have been optimized to take advantage of the high-performance and low-power features of Mythic Analog Matrix Processors (Mythic AMPTM). Developers can focus on model performance and end-application integration instead of the time-consuming model development and training process. Available pre-qualified models in development:
Object Detection and Classification YOLOv3, YOLOv5, ResNet-50, ResNet-18
Object Detection and Classification YOLOv3, YOLOv5, ResNet-50, ResNet-18
Scene Segmentation SegNet
Human Pose Estimator OpenPose Body25

Resources

Mythic Whitepaper / Blog

Download

MM1076 and ME1076 M.2 Cards Product Brief

Download

Mythic Whitepaper / Blog

Download
Return to Portfolio
Product Inquiry

The post MM1076 M.2 M Key Card appeared first on Mythic.

]]> ME1076 M.2 A+E Key Card https://mythic.ai/products/me1076-m-2-ae-key-card/ Wed, 08 Dec 2021 18:54:33 +0000 https://mythic.ai/?post_type=product&p=132 The ME1076 M.2 A+E key card enables high-performance, yet power-efficient AI inference for edge devices and edge servers.

The post ME1076 M.2 A+E Key Card appeared first on Mythic.

]]>

ME1076 M.2 A+E Key Card

Overview

The ME1076 M.2 A+E key card enables high-performance and power-efficient AI inference for edge devices and edge servers. The M.2 card’s compact form-factor and popularity makes integration into many different systemsstraightforward. The ME1076 is designed with the M1076 Mythic AMP™ which is arranged in an array of AMP tiles each featuring a Mythic ACE™. The ME1076 is ideal for processing deep neural network (DNN) models in a variety of applications, including video surveillance, industrial machine vision, drone, AR/VR, and edge servers.

ME1076 M.2 A+E key card

Features

M1076 Mythic AMP™ with support for up to 80M weights on-chip

No external DRAM required

SMBus for EEPROM and PMIC access

Pre-qualified networks including object detectors, classifiers, pose estimators, with more being added

OS Support: Ubuntu, NVIDIA L4T, and Windows (future release)

Model parameters stored and matrix operations executed on-chip by AMP tiles

2-lane PCIe 2.1 for up to 1GB/s bandwidth

Support for standard frameworks, including PyTorch, TensorFlow 2.0, and Caffe

Small 22mm x 30mm form factor

Workflow

DNN models developed in standard frameworks such as Pytorch, Caffe, and TensorFlow are implemented and deployed on the Mythic Analog Matrix Processor (Mythic AMPTM) using Mythic’s AI software workflow. Models are optimized, quantized from FP32 to INT8, and then retrained for the Mythic Analog Compute Engine (Mythic ACETM) 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

Models

Mythic provides powerful pre-qualified models for the most popular AI use cases. Models have been optimized to take advantage of the high-performance and low-power features of Mythic Analog Matrix Processors (Mythic AMPTM). Developers can focus on model performance and end-application integration instead of the time-consuming model development and training process.  Available pre-qualified models in development:
Object Detection and Classification YOLOv3, YOLOv5, ResNet-50, ResNet-18
Object Detection and Classification YOLOv3, YOLOv5, ResNet-50, ResNet-18
Scene Segmentation SegNet
Human Pose Estimator OpenPose Body25

Resources

Mythic Whitepaper / Blog

Download

MM1076 and ME1076 M.2 Cards Product Brief

Download

Mythic Whitepaper / Blog

Download
Return to Portfolio
Product Inquiry

The post ME1076 M.2 A+E Key Card appeared first on Mythic.

]]>