Sold by: Degas
Deployed on AWS
A geospatial foundational model that can be fine-tuned to your specific earth observation tasks.
Overview
Degas 100M is a pretrained cutting-edge geospatial foundational model designed for highly accurate earth observation. It uses an in-house developed spatio-temporal SwinMAE architecture to produce a robust geospatial model that can seamlessly adapt to custom downstream tasks. It handles both single capture imagery and multi-timestamp (3 steps at most) data as input, making it a reliable option both for land cover analysis and temporal change detection.
Highlights
- Easy to finetune: We offer a simple interface that automates the process of training and deploying the model via SageMaker. With just a Jupyter notebook you can easily finetune the model into you specific tasks and quickly deploy it into production.
- Input versatility: Designed for straightforward implementation into your existing workflows. It handles 3+ channels for both single capture imagery and multi-timestamp captures.
- High performance and cost reduction: Degas 100M demonstrates superior performance when compared to state-of-the-art geospatial foundational models. We demonstrated improvements on flood mapping, wildfire scar mapping and land cover classification. It particularly excels at the latter, with a 10.4% accuracy improvement on the PhilEO benchmark dataset. More details on the technical paper or contact us directly: sales@degasafrica.com.
Details
Sold by
Delivery method
Latest version
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.p3.2xlarge Inference (Batch) Recommended | Model inference on the ml.p3.2xlarge instance type, batch mode | $1.00 |
ml.p3.2xlarge Inference (Real-Time) Recommended | Model inference on the ml.p3.2xlarge instance type, real-time mode | $1.00 |
ml.p3.2xlarge Training Recommended | Algorithm training on the ml.p3.2xlarge instance type | $1.50 |
Dimensions summary
You pay by the hour based on the work you run with the Degas 100M geospatial model. All three options use the same ml.p3.2xlarge instance type, so pricing varies by task, not instance size. Two options cover inference: batch mode for processing grouped data, and real-time mode for immediate responses. The third option covers training, where you adapt the model to your own data. You can use any option on its own. Usage-based billing means your cost scales with host hours consumed per task.
Top-of-mind questions for buyers
Every option runs on one ml.p3.2xlarge instance, a GPU-backed instance type. You pay per host-hour, meaning each hour one instance runs counts as one billable unit. The three options differ by task, not hardware, so the underlying compute is the same across all of them.
Batch inference processes grouped data in scheduled runs, so you pay host-hours only while a batch job runs. Real-time inference keeps the model ready to answer requests on demand, so you pay host-hours for the time it stays active. Each meters running time independently.
Charges accrue per host-hour while an instance runs. If no training or inference job is active and the instance is stopped, software charges stop. Underlying AWS infrastructure fees, such as storage, may still apply separately. Each of the three options meters only its own running time.
www.degasafrica.com
Helpful?
Vendor refund policy
Degas does not offer refunds in any cases.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Amazon SageMaker algorithm
An Amazon SageMaker algorithm is a machine learning model that requires your training data to make predictions. Use the included training algorithm to generate your unique model artifact. Then deploy the model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
Initial release of Degas 100M
Additional details
Inputs
- Summary
Example input(s): Here you can see example datasets: burn scars and multi temporal crop classification .
- Input MIME type
- application/x-npy
https://huggingface.co/datasets/ibm-nasa-geospatial/hls_burn_scars
https://huggingface.co/datasets/ibm-nasa-geospatial/hls_burn_scars
Resources
Vendor resources
Support
Vendor support
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
CTrees' global aboveground biomass (AGB) estimates, annually for 2000-2025 at a standard spatial resolution of 100 meters.
Distributed Energy Generation Market Size, Trends and Insights By Technology (Wind Turbine, Solar Photovoltaic, Reciprocating Engines, Fuel Cells, Gas & Steam Turbine), By Application (Residential, Commercial & Industrial), and By Region - Global Industry Overview, Statistical Data, Competitive Analysis, Share, Outlook, and Forecast 2025 - 2034
The Saildrone Arctic 2019 dataset presents a unique collection of high-quality, near real-time, multivariate surface ocean, and atmospheric observations obtained through the deployment of Saildrone, an innovative wind and solar-powered uncrewed surface vehicle (USV). Saildrone is capable of extended missions lasting up to 12 months, covering vast distances at typical speeds of 3-5 knots and operates autonomously, relying solely on wind propulsion, while its navigation can be remotely guided from land. The 2019 Saildrone Arctic campaign featured six Saildrone USVs (jointly funded by NOAA and NASA) deployed during a 150-day cruise in the Bering and Chukchi Seas, spanning from 14 May 2019 to 11 October 2019. The primary mission objective for 2019 was to gather comprehensive atmospheric and oceanographic data in Alaskan arctic waters, which could lead to significant improvements in modeling of diurnal warming and understanding of the marginal ice zones. Additionally, these new data wi[...]
Denmark Consumer Segmentation provides a powerful set of consumer classification systems developed to assess the socio-economic and geodemographic portraits of neighborhoods across Denmark. It consists of analysis variables and two segmentation systems including international and market optimized categories. The data is based on census, government and commercial data.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.