Zcash NU7 has been live since Friday: the privacy coin’s network upgrade activated at block 3,543,405, mined on November 6, 2026 at 02:34 UTC. According to the deployment specification ZIP 259, NU7 cuts the target block time from 75 to 25 seconds, introduces per-block limits on shielded actions, activates the Network Sustainability Mechanism (NSM) and removes 60 percent of transaction fees from circulation. At the same time, version 4 transactions become invalid, with an unpleasant side effect: coins left in the oldest privacy pool, “Sprout”, can no longer be moved.
NU7 is not a spectacular feature leap. It is more of a maintenance and economics upgrade that brings faster confirmations, leaves the money supply unchanged and cuts off legacy baggage. That is exactly why it deserves a closer look, because a fair amount changes for wallets, exchanges and node operators.
Zcash NU7: what changes technically
25-second blocks (ZIP 218). Instead of every 75 seconds, a block should now arrive every 25 seconds, roughly 3,456 instead of 1,152 blocks a day. Mining difficulty is not reset; it adjusts at its normal rate, so actual block times approach the target only gradually. According to the ZecStats tracker, difficulty has already dropped substantially since activation. ZIP 218 estimates that the throughput ceiling for simple Orchard transactions rises from about 2.9 to 6.6 transactions per second.
Same issuance per day. So that three times as many blocks do not create three times as many new ZEC, the block subsidy drops from 1.5625 to 0.52083333 ZEC. Daily issuance therefore stays practically the same, and so does the 21 million ZEC cap. The third halving moves from block 4,406,400 to block 6,132,390 on paper, but because of the faster block cadence its date stays about the same, around late November 2028, according to ZecStats.
Limits on shielded actions. More blocks mean more data that wallets have to scan while syncing. NU7 therefore caps the number of shielded actions per block. According to ZecStats, that is at most 330 shielded actions in total, 330 each for Orchard and Ironwood, and 300 Sapling inputs plus outputs. The goal is to stop anyone from flooding the chain with transactions and forcing every wallet to do more work.
Longer expiry window. The transaction expiry delta grows from 40 to 120 blocks, which keeps the wall-clock window roughly the same as before.
Network Sustainability Mechanism: 60 percent of fees
The most interesting economic change is the NSM (ZIPs 235 and 237). It tackles a problem Zcash inherited from Bitcoin: with each halving the block subsidy shrinks, and it is an open question whether fees alone will give miners enough incentive to secure the network in the long run.
From NU7, the miner receives only 40 percent of a block’s transaction fees. The remaining 60 percent is removed from circulation and recorded in a publicly visible NSM balance. It is not burned: under ZIP 237, these funds will be gradually reissued through future block rewards starting at block 8,939,664, expected around February 2031, on top of the existing halving schedule and within the 21 million cap. The NSM balance is consensus state, but not a value pool of its own, and it adds no new transaction-level data.
In practice the effect is small today, because fees per block are low. ZecStats puts them at around 0.00175 ZEC per block. The mechanism is more about setting the course for a time when the subsidy barely matters anymore.
The end of version 4, and of Sprout
With NU7, all version 4 transactions become invalid, not just Sprout transactions. According to ZIP 2003, that also covers transparent-only and Sapling transactions as well as offline-signed transactions in that format. Versions 5 and 6 remain valid. Because Sprout transfers can only be represented in version 4 transactions, the remaining Sprout balances have been unspendable since activation.
Data from the ZecStats NU7 tracker puts the pool at about 12,224.21 ZEC. The distinction matters: the coins are frozen, not burned. ZIP 2003 leaves open that a future upgrade could restore access, but explicitly warns that this may never happen. No concrete recovery path has been approved so far.
For users, this means: anyone who still had funds in Sprout and did not migrate them before activation currently cannot reach them. For everyone else, the key point is that wallets and services must stop building version 4 transactions. ZIP 259 notes that simply re-signing a version 4 transaction does not make it valid.
Who needs to act
- Node operators: According to ZecStats, Zebra 7.0.0 or Zakura 1.7.0 and later are required; zcashd does not support NU7. Setups that need the zcashd interface can use a compatibility sidecar. Per ZIP 259, nodes that were not upgraded may end up following a pre-NU7 branch.
- Wallet users: Update your wallet app. For Zodl, ZecStats names version 3.19.0 or later as a requirement. ZecStats says it has not verified NU7 statements for hardware wallets such as Ledger or Keystone, so check the vendor’s guidance.
- Exchanges and payment services: Three times as many blocks per hour. How many confirmations a service requires remains its own decision. Version 4 transactions from legacy tooling are invalid.
- Wallet and SDK developers: Stop creating version 4 transactions, account for the 120-block expiry and stay under the per-block shielded limits.
What is not in NU7
What is missing is just as telling. According to ZIP 259, NU7 does not deploy ZIPs 233 and 234, so there is neither voluntary removal of funds via the NSM nor issuance smoothing. Zcash Shielded Assets (ZSAs), memo bundles and the longer-term projects Crosslink and Tachyon are not part of the upgrade either, and there is no new transaction format. The specification justifies the deliberately narrow scope by saying NU7 should only contain changes with strong support from coinholders, developers and ecosystem organisations. Another format change right after version 6, introduced with NU6.3, would have burdened parsers and wallets again.
There is also an exception in the schedule: for NU7, the usual rule that about three months should pass between the first supporting software release and activation was waived. Deployment communications therefore had to explicitly warn operators to upgrade in time.
Putting it in perspective
NU7 follows a pattern visible across many chains right now: shorter block times for better everyday usability. Base has just moved to 200ms blocks with Denim, and Solana is working on millisecond-level finality with Alpenglow. At 25 seconds Zcash plays in a different league, but it noticeably shortens the wait for private payments: a merchant sees the first confirmation after about 25 seconds on average instead of 75.
The most controversial part remains Sprout. Technically, switching off version 4 is well justified, and Sprout has been considered legacy for years. For the holders of a little over 12,000 ZEC, “frozen, perhaps forever” is still a harsh cut. As with every consensus change, upgrades are not just engineering but governance. Whether and how the community decides on a recovery path is the open question after NU7.
