Vendor review · licensed marketplace

Opendatabay

A young UK marketplace for buying and selling AI-ready datasets — breadth and self-serve convenience, with a licensing-first pitch still being proven.

The verdict. Useful for teams that want to browse off-the-shelf, synthetic, or open datasets without a sales cycle; per-listing rights depth is the thing to check before relying on it for media training data.
  • Self-serve breadth - open, synthetic, and premium datasets across many modalities in one storefront, with instant access
  • Published fee schedule - a rare bit of pricing transparency; the tiered 5–30% seller commission is public
  • Licensing-first framing - seller verification and rights-cleared positioning, uncommon for a general data marketplace
  • Young company - recently registered in the UK; large-buyer track record and catalog depth are still building
  • Provider-set licences - rights and consent depth vary listing by listing, with no file-level provenance standard
  • Open/synthetic weighting - headline dataset counts include free open data, so scarce premium supply is thinner than it looks

What Opendatabay is

Opendatabay Ltd, a recently registered company in England and Wales, runs a self-serve data marketplace where providers list datasets — open, synthetic, and premium — for AI and LLM training, alongside an open-data repository and a dataset-request service.

Modalities and languages they cover

Listing-dependent rather than curated: the company markets text, image, audio, video, code, and agentic-trajectory data, plus large open and synthetic collections. There is no fixed language or modality catalog — coverage is whatever sellers have listed.

Sourcing model

Third-party aggregation. Data comes from outside providers who, per the company, pass seller verification and due diligence before listing; Opendatabay itself does not commission or capture data.

Rights & provenance posture

Opendatabay says it focuses exclusively on licensed, consented, rights-cleared datasets and verifies that sellers have the legal authority to sell. Verification is described at the seller level, though — licence terms are set per listing, and public materials do not show file-level consent or provenance documentation, so audio/video buyers should request the underlying records asset by asset.

Pricing

Self-serve: open datasets are free and premium datasets are priced by their sellers. Unusually, the seller fee schedule is public — a tiered commission from 30% on sales under £100 down to 5% above £50,000.

Where Opendatabay falls behind

Trust is asserted at the storefront level, not carried by the file. A meaningful share of the catalog is open-repository or synthetic data available elsewhere, premium listings carry provider-written licences of varying depth, and there is no standardized consent record travelling with the media — the gap a provenance-first licensor is built to close.

Sources

Alternatives

Frequently asked questions

Is Opendatabay data rights-cleared for AI training?

The company says it lists only licensed, consented datasets and verifies sellers’ authority to sell. Licence terms are still set per listing, so confirm AI-training and redistribution rights on each dataset before buying.

How much does Opendatabay cost?

Open datasets are free; premium datasets are priced individually by sellers. On the sell side, Opendatabay takes a published tiered commission of 5–30% depending on sale price.

Who is Opendatabay best for?

Teams that want quick, self-serve access to off-the-shelf or synthetic datasets, and data owners wanting a low-friction storefront. Bespoke, consent-documented media capture is not the model.

Want data you can actually defend in diligence?

fiund licenses real-world audio and video at the source, with the rights cleared before anything moves.

Send a brief