Saves the drop's items from files uploaded with one upload_batch_id through POST /api/v2/drops/{slug}/items/media, naming them by filename instead of by media token. The request stays small for large drops: a 15,000-item save is about a tenth of the size of the same save by media token, and it is validated with one storage listing instead of one lookup per file. It behaves exactly like POST /api/v2/drops/{slug}/items/media/save otherwise: each save replaces the drop's items, a manifest from POST /api/v2/drops/{slug}/items/manifest supplies token ids and metadata, and without one items are numbered 1 to n in the order of filenames.

Returns 400 when a filename is listed twice or was not uploaded in the batch, naming the files concerned. If a file was uploaded more than once in the batch, the most recent upload is saved. Not available on ERC-1155 self-mint drops.