Check out the newest way to compare different models for a task/agent harness: AutoEvals

Command Reference

Uploads

Inspect inference uploads — status, list, and download URL.

Inspect inference uploads you've staged: check their status, list the uploads backed by a project, and get a presigned URL to download the raw payload. The chunked upload write flow itself stays in the SDK.

Commands

Run inf upload <verb> --help for each verb's options. The generated surface covers:

VerbDescription
listList a project's inference uploads
getGet an inference upload's status
download-urlGet a presigned download URL for an inference upload
initiate/complete/abortBegin, finish, or abort a chunked (R2 multipart) upload
archive / unarchiveArchive (soft-delete) or restore an inference upload
  • inf dataset materializes uploads (and other sources) into eval/training datasets.

On this page