Datasets
Buyer room
BOL.AIManage supplier listing
Bol.ai — Document-AI corpus plus derived shipment ledger: real-world Bills of Lading, commercial invoices, packing lists and CMR waybills (PDF/scan/photo) paired with verified structured extractions across 20+ fields, plus the cross-document reconciliation results that flag where those documents disagree. Small vendor, but the artifact is unusually dense: every document yields a verified 20+ field record with its source image and reconciliation verdicts, across many distinct carrier templates — the layout diversity that is expensive for a buyer to assemble independently. Absolute volume is lo…
supplier index read from their site
Document-AI corpus plus derived shipment ledger: real-world Bills of Lading, commercial invoices, packing lists and CMR waybills (PDF/scan/photo) paired with verified structured extractions across 20+ fields, plus the cross-document reconciliation results that flag where those documents disagree
Scale est. low tens of thousands of documents to dateFrom 2 yearsCoverage Industrials
Sample, licence terms, pricing and eval results when Bol.ai publishes them. Until then, discover alternatives today with a 7-day trial.
Start 7-day trial →What buyers ask about Bol.ai, answered from this page.
Bol.ai — Document-AI corpus plus derived shipment ledger: real-world Bills of Lading, commercial invoices, packing lists and CMR waybills (PDF/scan/photo) paired with verified structured extractions across 20+ fields, plus the cross-document reconciliation results that flag where those documents disagree. Small vendor, but the artifact is unusually dense: every document yields a verified 20+ field record with its source image and reconciliation verdicts, across many distinct carrier templates — the layout diversity that is expensive for a buyer to assemble independently. Absolute volume is lo…
Bol.ai offers (Alternative, Supply Chain, Reference) — est. low tens of thousands of documents to date, each yielding 20+ labelled fields (bounded from a recent launch on per-document pricing with no disclosed customer count); every processed document retains its source image, so the image-to-JSON pair count equals document count.
Training and evaluating document-AI / OCR extraction models on genuinely heterogeneous carrier layouts (the hard part of the problem, since every carrier's B/L template differs); document-type classification; cross-document reconciliation and data-validation models; entity resolution across shipper/consignee spellings; derived freight and trade-flow analytics on lane, weight, volume and Incoterms
The data is with 2 years of history.
Coverage spans APAC, Europe, US, Japan; Marine Shipping, Air Freight & Services; alternative, supply_chain, reference; equities.
Bol.ai has not added their own details yet. Not yet on file: