Model Approval Process

While building models for gravityAI, one must always remember that buyers are trusting you (and, in extension, us) to provide a solution that is reliable, performant, well documented, and secure. In order to achieve some sense of this, models must go through a process to get approved and listed in our catalog:

Namely:

1. Models must be well documented. Note: If you want an example of a well documented model, download the documentation from this one HERE

   • What’s the input and output?
   • What are the use cases of your model?
   • What are the ethical concerns of your model (if any)?
   • What are the limitations of your model?

2. Models must be performant and reliable.

   • How much memory should be allocated to Docker in order to run your model?
   • What specs are required to run the model?
   • What is the expected runtime for your model per job? How does this runtime scale with varying input sizes?
   • How accurate is your model?
   • How precise is your model?

3. Models must be as secure as possible.

   • Vulnerabilities will be scanned for and should there be vulnerabilities of a critical level, they will not be accepted. So try to make sure that the model is using up-to-date libraries, if at all possible.

In particular, from the point you list your model, the approval process might flow as follows…

Model Approval FAQ

Q: Why hasn’t my model been approved yet? It’s been ___ weeks.
A: While we endeavor to get to as many models as possible, we are a small team with a lot of people wearing different hats. Sadly, it might take a while before we can get to your particular model. But get to your model we shall!