The AWS Panorama Appliance brings computer vision to your existing, on-premises industrial cameras. You purchase the physical device up-front, and then are charged based on the number of unique camera video streams processed on the Appliance each month. You’re also charged for cloud storage of assets deployed to the Appliance (including ML models and business logic).
- Appliance: You pay for the Appliance up-front, as a one-time cost.
- Appliance usage: You are charged monthly based on the number of unique camera video streams processed on the Appliance at any time in a given month. You are charged $8.33 per month per camera stream.
- Appliance Developer Kit usage: There is no charge to use the AWS Panorama Developer Kit. It is not for production use and is strictly for application development in non-production environments.
- AWS Panorama cloud storage: AWS Panorama stores versioned copies of all assets deployed to Appliances (including ML models and business logic) in the cloud. You are charged $0.10 per-GB, per-month for this storage.
Summarized Appliance pricing is below:
|Panorama Appliances||Appliance cost||Appliance usage|
|AWS Panorama Appliance Developer Kit||$2,499 per device||$0 (free for the Appliance Developer Kit)|
|AWS Panorama Appliance||$4,000 per device||$8.33 per month per active camera stream|
You may incur additional charges if the business logic deployed to your AWS Panorama Appliance uses other AWS services. For example, if your business logic uploads ML predictions to Amazon S3 for offline analysis, you will be billed separately by Amazon S3 for any storage charges incurred. For details on pricing for other AWS Services, see the pricing section of the relevant AWS Service detail pages.
During Preview, the AWS Panorama cloud storage fees are free; after Preview, you will be charged $0.10 per-GB, per-month. All other charges are the same during Preview.
In this example, you purchase an AWS Panorama Appliance to improve workplace safety. You want to monitor if pedestrians are crossing into heavy machinery operating zones in your facility. You train a machine learning model in Amazon SageMaker to identify when people come within the restricted machinery operating zone. Your write business logic that logs these occurrences and sends results from the Appliance to your site safety system in real-time.
Your workplace safety artifacts, which include the ML model and business logic, are 200MB in size. You run this workplace safety model on your Appliance against seven camera streams continuously.
In this example, you want to automate the quality inspection of parts moving along the manufacturing lines at five sites. You purchase one AWS Panorama Appliance for each site (5 total), where there are 10 cameras already installed per site (50 cameras total). You use an existing ML model you previously trained in Amazon SageMaker to identify when parts have common manufacturing errors. Your associated business logic sends results from the Appliance to the site manufacturing analytics system.
Your quality inspection artifacts, which include the ML model and the business logic, are 2GB in size. You use your AWS Panorama run this quality inspection model against all 10 camera streams for each site’s Appliance (10 cameras x 5 appliances = a total of 50 camera streams).
You also purchase one AWS Panorama Appliance Developer Kit to test updates to your ML model and business logic using non-production camera streams. You publish an updated version of the quality inspection ML model every 3 months, but only store the latest two versions in your account. So, for 2 versions at 2GB each = 4GB total AWS Panorama cloud storage.