D · Digital Products · Data & Datasets
Licensed AI training datasets
Text, image, audio and structured datasets licensed for training, fine-tuning or evaluating AI models. Provenance and training rights are the deciding factors: confirm where every record came from, how it was labeled and that the license explicitly covers model training.
New market — be the first to list
More in Data & Datasets
0 results
No live listings in this market yet
This market is open for listings. Sellers can list here today, and buyers can post a request so sellers know what's wanted.
Buyer guide
- Get explicit written training and commercial-use rights, not just a general data license.
- Review label quality on a sample and ask how disagreements between annotators were resolved.
- Confirm personal data has been excluded and copyrighted material is licensed.
What to compare
- 01Modality and domain
- 02Provenance and consent
- 03Labeling method and quality checks
- 04Training and commercial rights
- 05Format and splits
Use cases
- Fine-tuning a domain-specific model
- Building evaluation sets
FAQ
What should an AI training dataset license include?
It should state that model training and commercial use of the resulting model are permitted, and disclose the data's provenance. The Node Passport records these terms.