Join the Starter Model Challenge and submit AI models with the datasets, preprocessing, model architecture and instructions that developers need to turn them into production-ready Edge AI models. Read more about how to join and what you can win below!

Join our Starter Model Challenge to solve exciting problems and become part of the DEEPCRAFT™ Edge AI model maker community!

What are Starter Models?

DEEPCRAFT™ Starter Models are open-source models that include the datasets, preprocessing, model architecture and instructions that developers can utilize as basis to train production-ready Edge AI models with the development platform DEEPCRAFT™ Studio. Starter Models cover a wide variety of use cases and sensors, and are ready to be refined and trained based on your needs. The purpose is to inspire developers and to help kick-start your Edge AI journey. Read more about Starter Models.

Why should you participate?

  • Get your model highlighted on the Starter Model page of our website for a month.
  • Get a bag of Imagimob and DEEPCRAFT™ merch
  • Be part of the open-source community of DEEPCRAFT™ model makers we are building! 

How do you win?

The Starter Model challenge will have two winners:

Winner 1
Our team will choose from all the Starter Models submitted via our GitHub up to and including September 30, 2025. We are looking for a model that has a real-world application or solution. It should show promise for further development, with enough data to make interesting conclusions and observations (including negative data), and be sufficiently documented so that others could use it in their development
Open until end of September, winner announced in October

Winner 2
The Starter Model (excluding those developed by Imagimob) most-used by DEEPCRAFT™ Studio users will be our second winner. This will be through the rest of 2025, but the sooner you submit, the more time you have to get those users and win in this category!
Winner announced in January 2026

Are you ready?

You'll find information all the information you need to get started on our GitHub