ZetaChain accumulated hundreds of millions of transactions before deciding its Layer 1 was no longer the best place to spend resources. The numbers underneath that decision tell a more revealing story than the wind-down itself.
ZetaChain is preparing to do something few Layer 1 networks willingly plan for: make its own blockchain disappear.
On September 20, holders approved Proposal 68, setting ZETA on a path toward becoming a native Solana token while ZetaChain’s Cosmos-based Layer 1 is eventually wound down. The vote was overwhelming. According to The Block, 99.4% of participating voting power supported the proposal, with 58% participation, well above the required quorum.
But the vote does not shut anything down immediately. ZetaChain’s existing network remains live, and another governance proposal still needs to define the snapshot, migration mechanics, exchange arrangements, and the final halt.
The more interesting question comes before any of that.
How does a blockchain that accumulated well over 247 million transactions reach the point where its team no longer sees maintaining an independent L1 as the best place to spend resources?
The answer lies in the difference between activity accumulated over years and economic activity retained today.
247M Transactions, but What Stayed?
ZetaChain was not launched quietly.
When the project announced a US$27 million funding round in August 2023, its stated mission was to accelerate what it called the “first public EVM-compatible L1 blockchain that connects everything.”
The pitch was ambitious: Bitcoin, Ethereum and other networks could become accessible through a single interoperability layer, reducing the need for conventional bridges and wrapped assets. ZetaChain said its testnet had already attracted 1.7 million users and processed well over 13 million transactions before the mainnet.
Early mainnet numbers appeared to validate that momentum.
In May 2024, ZetaChain reported well over 100 million Universal EVM transactions in roughly the first 90 days after the mainnet beta.
That works out to more than 1.1 million transactions per day on average during that early period.
Fast-forward to 2026, and the cumulative number still looks impressive. A recent ZetaScan explorer snapshot showed around 247.3 million total transactions and 12.2 million addresses.
Yet the same snapshot showed just 687 daily transactions and 0.01% network utilization.
Those figures should not be interpreted as a perfect apples-to-apples comparison. Launch campaigns, incentives, applications and transaction patterns change over time. A lifetime transaction count also cannot tell us how many transactions represented recurring economic users.
But the scale of the divergence is difficult to ignore.
Ravel calculates that 687 daily transactions represent only around 0.062% of the roughly 1.11 million-per-day pace implied by ZetaChain’s first 100 million transactions.
Put differently, present throughput in that snapshot was roughly 99.94% below the early mainnet pace.
The question is not whether those original transactions were “real.”
It is how much of that activity became behaviour users continued once the novelty, incentives and early ecosystem push faded.
The L1 Security Economy Has an Even Bigger Gap
Transaction counts tell only part of the story. Fees make the economics harder to miss.
At the time of writing, DefiLlama data showed only about US$31.10 in ZetaChain chain gas fees over the previous 30 days.
Annualized mechanically, that would equal roughly US$378 per year.
That number can then be compared with one of the largest remaining pieces of the current ZetaChain economy: staking.
ZetaChain’s own State of Private AI dashboard reported 455.1 million ZETA staked, or 21.67% of total supply, as of September 23.
Its reward calculator used an estimated 9.4% annualized reward rate.
Ravel’s calculation:
455.1M ZETA × 9.4% = 42.78M ZETA in annualized rewards (rounded)
Using the dashboard’s US$0.0514 reference price, that represents roughly US$2.2 million in annualized notional staking rewards.
Compared with an estimated US$378 in annualized gas fees, the resulting gap comes to:
5,800x
That does not mean ZetaChain is spending US$2.2 million in cash to earn US$378.
Staking rewards are token-denominated incentives, not the same thing as disclosed operating expenses, and this calculation says nothing about the actual infrastructure costs paid by validators or ZetaChain contributors.
Instead, it measures something narrower.
The value of the token incentives implied by the current staking system is thousands of times larger than the user gas fees currently being generated by the chain.
Call it the Security Subsidy Gap.
Viewed from that angle, retiring the L1 starts looking less like an inexplicable retreat and more like a decision to stop maintaining an independent security layer whose observable fee demand has become extremely small.
The Chain Can Disappear, The Tokenomics Don’t
The transition also creates an unusual split between ZetaChain’s infrastructure and its token.
Under ZetaChain’s Solana migration plan, native ZETA is expected to move to Solana on a 1:1 basis as an SPL token. Total supply stays unchanged, existing vesting schedules remain on their original dates, and ZETA already issued on Ethereum and BNB Chain is outside the proposal.
What disappears is the underlying network.
ZetaChain’s own validators would eventually be replaced by Solana’s validator infrastructure, while contributors redirect their focus toward ZETA, the Anuma AI application and an application layer built on Solana.
Even the future of staking is not finalized. ZetaChain says how staking and staking rewards work after the migration is still being explored.
Meanwhile, the existing supply schedule continues.
CoinGecko lists an October 1 unlock of 44.26 million ZETA, equivalent to about 2.1% of total supply. The allocation includes 13.13 million ZETA for core contributors, 12.83 million for the Protocol Treasury and 9.33 million for purchasers and advisers, among other categories.
That creates a less obvious version of the wind-down story:
The infrastructure thesis is being retired, but its original supply schedule is not.
There is nothing inherently unusual about honoring established vesting commitments after a strategic pivot.
The harder question is what future demand replaces the utility ZETA previously derived from gas payments, validator staking and securing an independent blockchain.
ZetaChain believes the answer is AI.
Anuma Is Real Usage, Value Capture Is Still the Test
This pivot is not starting from zero.
ZetaChain has spent 2026 building Anuma, a private multi-model AI application centered around portable encrypted memory. Every Anuma account creates a wallet.
By September 22, ZetaChain reported 306,875 wallets created, 1.27 million inference requests and 34.4 billion AI tokens processed since February. Its research dashboard also tracked roughly 293,000 conversations over the latest 90-day window.
Those are meaningful product metrics.
But they need the right denominator.
Dividing 1.27 million lifetime inference requests by 306,875 created wallets yields an average of 4.1 requests per wallet since launch.
That is not an estimate of actual requests per active user. Individual usage is unlikely to be evenly distributed, dormant wallets remain in the denominator, and heavy users could generate many more requests than everyone else.
It does show why wallets created cannot substitute for retention.
What Ravel still cannot see publicly are Anuma’s monthly active users, weekly active users, retention cohorts, paying subscribers or revenue.
Those figures matter because the entire ZETA value-capture model is being redesigned around application demand.
Today, users can lock ZETA for AI credits, and locking 100,000 ZETA unlocks Anuma Pro. ZetaChain also says ZETA can eventually support model access, encrypted memory, agent activity, permissions and settlement.
Its stated philosophy for the transition is straightforward: “Products people actually use come first, with ZETA demand growing alongside that usage.”
That is also the cleanest way to judge whether this pivot works.
Not by counting how many wallets migration brings to Solana.
Not by counting lifetime AI prompts.
But by measuring whether recurring Anuma usage causes users to continuously acquire, lock or spend ZETA at a scale large enough to replace the utility being removed with the L1.
ZetaChain’s Failure Wasn’t Necessarily Acquisition
Calling ZetaChain ‘simply a failed blockchain’ would miss the more interesting lesson.
It acquired users.
It processed hundreds of millions of transactions.
It accumulated millions of addresses.
At one point, it was processing transactions at a pace exceeding one million per day.
What the current data suggests is that ZetaChain struggled with something harder:
conversion.
Conversion of transactions into durable network usage.
Conversion of addresses into retained users.
Conversion of ecosystem activity into fees.
And ultimately, the goal is to convert the interoperability thesis into sufficient economic demand to ensure that owning and securing sovereign blockspace remains necessary.
There is also a credible counterargument.
If the differentiated product ZetaChain now wants to build is an AI application rather than Blockspace itself, continuing to maintain an independent Cosmos-based validator network may be unnecessary complexity. Solana already provides security, liquidity, wallets and payment infrastructure. Moving there could allow ZetaChain to devote more engineering resources to the layer users actually interact with.
The current chain data arguably makes that argument stronger.
But it does not yet prove the next thesis.
ZetaChain’s first experiment accumulated 247 million transactions before its own L1 became expendable.
Its next experiment is more demanding.
Anuma now has to prove that usage can become durable economic demand and that ZETA can retain value when the blockchain it was originally built to secure no longer exists.
That is the metric worth watching after the hype.
