What is Tezos (XTZ)?

Every other blockchain upgrades by hard fork: developers propose a change, the community argues, and if the argument goes badly enough the chain splits. Bitcoin's block size war produced Bitcoin Cash. Ethereum's DAO decision produced Ethereum Classic. Coordination failure is a permanent risk in the design.

Tezos made upgrading a protocol function. Anyone can propose an amendment; bakers vote across several periods; if it passes it is tested on a temporary fork and then installed automatically on mainnet. No coordination, no split, no chance for a faction to walk away with a copy of the chain.

More than twenty upgrades have shipped this way and none has produced a contentious fork. That is a genuine achievement and it is Tezos's real contribution: the mechanism works, it has worked for years, and nobody else has adopted it.

Adaptive Issuance

The Paris upgrade of April 2024 replaced Tezos's fixed issuance with a dynamic one targeting a 50% staking ratio. When too little is staked the rate rises to attract more; as the ratio climbs toward target, inflation falls.

It worked in the direction intended. Direct staking went from around 7% of circulating supply when the mechanism launched to roughly 29% — a fourfold increase. That is an unusually clear case of a monetary policy change producing the behaviour it was designed to produce, and the chain is still short of its 50% target, which means issuance has not yet fallen to its floor.

The 2026 performance work

Tezos's twenty-first protocol upgrade, activated on 30 June 2026, raised Data Availability Layer bandwidth to 10 MB/s — a fifteenfold increase — and made the DAL attestation delay dynamic, cutting Layer 1 confirmation from roughly 66 seconds to 12–18. Etherlink, the EVM-compatible Layer 2, continues on its own upgrade track and gained native Ledger support in January 2026.

A 66-second confirmation was a real usability problem, and fixing it in 2026 is late — competitors solved latency years ago. It was fixed by an on-chain vote with no fork, which is the pattern for this chain: the right change, shipped cleanly, after the market moved on.

• Baking and staking to secure the chain, with both delegation and direct staking available.
• Governance — XTZ holders and bakers vote on every protocol amendment, and those votes actually install code.
• Gas for transactions on Layer 1 and, in part, for Etherlink.
• Collateral and liquidity across Tezos DeFi, which is small.

Tezos found durable niches in digital art and in institutional pilots, particularly in France and among European financial institutions, where formal verification and on-chain governance were selling points. Those are real footholds rather than announcements, and they are small markets.

XTZ has no supply cap. Issuance pays bakers and stakers, and since April 2024 the rate moves with the staking ratio under Adaptive Issuance, targeting 50% staked. At roughly 29% direct staking the mechanism is still pushing the rate up rather than down, which means current inflation is above where the design intends to settle.

The original 2017 fundraiser was one of the largest of its era and was followed by a well-documented governance dispute between the founders and the foundation that delayed the network's launch by roughly a year. The allocation from that period has long since distributed.

What Adaptive Issuance actually does for a holder

It makes the choice explicit. Stake, and you receive issuance that is being deliberately elevated to attract participants like you. Do not stake, and you are diluted by it. As the ratio approaches 50% the rate falls and the penalty for not staking falls with it. It is a coherent mechanism and it is a treadmill for anyone holding XTZ passively.

Tezos calls validators bakers, and there are two ways to participate, which changed materially with Adaptive Issuance.

Delegation is the older and gentler route: your XTZ never leaves your wallet, stays liquid, is never at risk, and you receive a share of your baker's rewards. Direct staking, introduced with the new model, locks the tokens with the baker for a higher reward — and exposes them to slashing if that baker misbehaves.

The distinction is the point of the design. Passive delegators earn less and risk nothing; direct stakers earn more and share the baker's risk. Anyone reading a single Tezos staking rate should establish which of the two it describes, because they are different products.

Right ideas, wrong decade

Self-amending governance, formal verification and on-chain upgrades are genuinely good ideas that Tezos implemented first and well. The market rewarded throughput, cheap fees and developer familiarity instead. Being early and correct has not translated into ecosystem size, and seven years is long enough to call that a pattern rather than a delay.

Ecosystem depth

Tezos DeFi is small, its developer base is small, and Michelson and the formal-verification toolchain are a real barrier for developers arriving from Solidity. Etherlink's EVM compatibility is the answer and it arrived into a crowded market of EVM layer 2s.

Inflation is currently elevated by design

Adaptive Issuance raises the rate until staking hits 50%. At roughly 29% the mechanism is still pushing it up, so holders who do not stake are being diluted at a rate deliberately set above equilibrium.

Governance turnout

On-chain governance is only as good as participation, and baker voting is concentrated among large operators. A mechanism that installs code by vote is a mechanism worth capturing, and Tezos's protection is the multi-period process rather than broad turnout.

Latency was fixed late

Confirmation times fell from about 66 seconds to 12–18 in June 2026. That is a large improvement and it arrives years after competing chains offered sub-second finality, and after the applications that needed it chose somewhere else.