What types of datasets are stored here?

Filecoin is particularly suited for storing large-scale datasets, although some smaller datasets can be found here as well. A broad variety of data is ​accessible via this site. This includes telemetry and imaging, historical archives, Creative Commons-licensed media, entertainment archives, scientific research papers, machine learning training sets and more.

How often is the Dataset Explorer updated?

The datasets are updated in real time, as on-chain deals are made on the Filecoin mainnet. You may see a 10 minute discrepancy between when a deal on the chain becomes sealed and active, and when it's visible on this site.

How do I retrieve a file or a dataset?

At the moment, retrievals must be done via lotus or other Filecoin tools like Powergate. Once you've found a file you'd like to download, click the code icon to generate a lotus command for its retrieval. Additionally, we're currently working on other retrieval methods that will require less tooling and overhead.

Where can I find a specific file from a dataset?

Browse a dataset or search files by filename in the search bar on any dataset's page. We're also working on building indices of datasets. These will give you the ability to see a tree of files located within each 32 GiB container, which will be especially useful for larger datasets containing numerous smaller files.

Can I propose a new dataset to be stored?

Yes, absolutely. Propose new datasets by creating an issue on Github. This is currently done under the Slingshot program.