National Research and Education Network
Education Cloud Data Center
Design & Simulation
Planning & Analytics
Enterprise Data Center
Server - Intelligent Computing
Enterprise Cloud Communications
Network Management System
Nine Steps to Response to the Pandemic Using Technology
Tackling the Emergency Together
Ubiquitous Connectivity Makes Digital Transformation Possible
Economical and Powerful AI Computing
Horizon Digital Platform
Connecting the Physical and Digital Worlds to Help Customers Achieve Successful Digital Transformation
253 Fortune Global 500 Companies Choose Huawei as Digital Transformation Partner
Secure And Resilient Solutions
Buy from Huawei
If you want to get more information about your project, you can submit your information and we will contact you as soon as possible.
If your company has signed an eDeal contract with Huawei, please buy your required product/solution via the link below.
Buy from resellers
Search for a nearby reseller and get direct contact information.
Become a Partner
Resources and Support
Huawei Authorized Learning Partner
Huawei ICT Academy
Huawei ICT Competition
Enterprise Training Service
Atlas 200 AI Accelerator Module
The Atlas 200 AI accelerator module (model: 3000) integrates the Ascend 310 AI processor to implement image recognition and image classification on the device side. Atlas 200 is widely used in AI scenarios such as intelligent cameras, robots, and drones.
• 22 TOPS INT8 in the size of half a credit card, supporting real-time analysis of 20-channel HD videos (1080p 25FPS)
• Multi-level computing power configuration: 22/16/8 TOPS
• Hibernation at milliwatts and wakeup in milliseconds, typical power consumption of 5.5 W, enabling edge AI applications
Supports pixel-level image segmentation and multiple algorithms, such as video analytics, posture identification, and vehicle recognition, and applies to scenarios such as intelligent security protection.
Industrial robots support precise visual control in manufacturing scenarios and home robots can identify faces, postures, gestures, and surrounding obstacles to implement customized functions such as meal delivery.
Supports real-time status identification for specified outdoor scenarios, for example, using drones to detect dangling or damaged power transmission lines.
|Parameters||Atlas 200 AI Accelerator Module
|AI Processor||Ascend 310|
|AI Computing Power||22 TOPS INT8
16 TOPS INT8
8 TOPS INT8
|Memory||LPDDR4X, 8 GB/4 GB, total bandwidth 51.2 GB/s|
|Encoding/ Decoding||• H.264 hardware decoding, 16-channel 1080p 30 FPS (2-channel 3840 x 2160 @ 60 FPS)
• H.265 hardware decoding, 16-channel 1080p 30 FPS (2-channel 3840 x 2160 @ 60 FPS)
• H.264 hardware encoding, 1-channel 1080p 30 FPS
• H.265 hardware encoding, 1-channel 1080p 30 FPS
• JPEG decoding: 1080p 256 FPS; encoding: 1080p 64 FPS; maximum resolution: 8192 x 4320
• PNG decoding: 1080p 24 FPS; maximum resolution: 4096 x 2160
|Port||• PCIe x4 Gen3.0
• 1 USB 2.0/USB 3.0
• 1 RGMII
|Interface Specifications||144-pin BTB connector|
|Typical Power Consumption||4 GB: 5.5 W
8 GB: 8 W
|Operating Temperature||-25°C to +80°C (-13°F to +176°F)|
|Dimensions (H x W x D)||8.5 mm x 52.6 mm x 38.5 mm (0.33 in. x 2.07 in. x 1.52 in.)|
Share link to：