This form is part of a research project at Duke University examining how AI models are publicly released under “open” terms.
We aim to identify patterns in openness, licensing, and transparency across different model releases.
This project is interested in answering two questions:
- How are “open” models being publicly released?
- What are the key trends in “open” AI model releases?
If you know of an open, partially open or model that claims to be an "open" AI model, please share its details below. Submissions will be reviewed and added to an open dataset for research purposes.
Updates on the project including code and datasets will be available online at sites.duke.edu/gabrieltoscano/