Buy & Swap Tagger (TAG)

Data work as the starting point

Tagger describes a permissionless system for collecting, labelling, managing and trading data used to train artificial-intelligence systems. Its basic premise is that useful machine-learning data still depends on human work and verification.

The documentation calls its reward process Proof-of-Human-Work. In this model, TAG is issued for completed data-labelling labour rather than for conventional computing-based mining.^{[1], [2]}

How TAG moves through the platform

The official documentation lists several uses for TAG: posting data tasks, staking, buying datasets, using datasets, subscribing to software and paying for model-customisation services.

This creates two sides of a marketplace. Organisations can fund work or buy data, while contributors can complete tasks and receive tokens. Whether an individual service is currently available should be checked in the live application.^{[3], [1]}

Supply and distribution

Tagger states a total supply of 405,380,800,000 TAG. Its published allocation assigns 300 billion to Proof-of-Human-Work, 85.3808 billion to Tag-to-Pump and 20 billion to liquidity.

Those amounts correspond to roughly 74%, 21.06% and 4.93%. They describe the project's allocation design, not the amount circulating at any given time.^{[4], [5]}

A staged emission model

The token-distribution page describes a halving coefficient. The first reduction is triggered after 50% of tokens have been issued, and the second after 75%, with further stages following the same progression described by the project.

This mechanism connects emissions to cumulative issuance rather than a simple calendar date. The documentation does not provide enough current issuance data to calculate the next event here.^{[6], [4]}

Governance and access

Tagger calls TAG both its native token and its governance token. Alongside voting, the token acts as the payment and incentive layer for work, data and software services.

The data marketplace indexes data and annotation files as NFTs, with ownership used to control access. That does not make every underlying dataset public; it is the access mechanism described by the project.^{[7], [8]}

What can and cannot be checked

The GitBook offers enough detail to explain the intended system, including an AI Copilot designed to help workers complete and verify labelling tasks.

However, documentation alone does not establish a current contract address or operational status of every feature. A careful user should treat those missing details separately from the documented design and avoid relying on third-party addresses.

The evidence is therefore strongest when describing Tagger's intended workflow and weakest where a reader needs operational verification. The documents explain who can earn TAG, what customers may spend it on, how supply is allocated and how later issuance is meant to slow. They do not, in the pages checked, let a reader independently confirm a current deployment address or audit. That difference should shape any practical decision. A detailed token model can be informative without being sufficient proof that a particular wallet asset is authentic. Before connecting a wallet, compare any address shown in an application against a current first-party announcement and check that the same network is named in both places. If Tagger restores an accessible main site or adds a contract page, that would close the largest information gap in this profile.^{[9], [3]}

What remains unknown

The project’s main website returned an access block, and the checked first-party documentation did not publish a contract address, audit report or clearly identified operating team. Those points are UNKNOWN and should be verified independently before using a contract.