# Zcash developers merge the first Tachyon component into the shared Zakura library

Project Tachyon lead Sean Bowe says the Udon release moves part of the proposed scaling design into production code aimed at more than 50,000 private payments a second.

- Published: 2026-09-30T06:00:33.939Z
- Canonical: https://polylog.news/crypto/2026-09-30/zcash-developers-merge-the-first-tachyon-component-into-the
- Publisher: Polylog (Crypto desk)
- Section: crypto
- Sources: [CoinDesk](https://www.coindesk.com/tech/2026/09/29/zcash-s-faster-private-payment-code-is-being-rebuilt-for-its-bigger-scaling-plan), [Cryptometer](https://www.cryptometer.io/news/zcash-advances-50000-tps-goal-with-tachyon-integration/)

Zcash developers released Udon, a component of the proposed Tachyon upgrade, into Zakura Common, the shared cryptography library used by the Zakura node software and other Zcash projects. Project Tachyon lead Sean Bowe confirmed the merge on Monday, [CoinDesk reported](https://www.coindesk.com/tech/2026/09/29/zcash-s-faster-private-payment-code-is-being-rebuilt-for-its-bigger-scaling-plan). Earlier work on the same library cut the time to create a private transaction from more than three seconds to under 200 milliseconds in some tests.

Tachyon's purpose is to reduce the data a node must download and verify as shielded usage grows, by aggregating proofs from many transactions into one. The stated long-term target is more than 50,000 private payments a second, [per coverage of the release](https://www.cryptometer.io/news/zcash-advances-50000-tps-goal-with-tachyon-integration/). That figure is a design goal, not a measured throughput. Tachyon has not been activated on the Zcash mainnet, and the remaining path runs through further development and security audits.

This design choice matters because of what it changes. Shielded systems historically forced each node to carry the cost of every private transaction, which set a firm limit on how many users a privacy chain could serve without centralizing verification onto a few powerful machines. Proof aggregation removes that limit directly, and doing the work in a shared library rather than a single client means wallets and other node implementations inherit the improvement rather than reimplementing it.

The risk is equally concrete. Aggregation changes what a verifier checks, and that is exactly where soundness defects arise, as Aztec's disclosure this week showed on a different network. Zcash is shipping the pieces into audited-in-advance library code rather than switching consensus first, which is the slower and more conservative order of operations.

## What this means

The mechanism is verification cost. Privacy chains scale only when the per-node burden of shielded traffic stops growing in step with usage, and proof aggregation is the method that changes that relationship. Exposed here are Zcash holders and shielded-pool users, whose network's usefulness for payments depends on whether throughput and proving time reach levels that consumer wallets can support, and by extension every privacy project that would copy the design. The precedent is the Sapling and Orchard upgrades, where circuit improvements shipped through libraries and clients well before a consensus activation, and the activation itself was uneventful. The likelier outcome over the next several months is a sequence of further Tachyon components landing in Zakura Common with measured proving-time improvements, and a mainnet activation debate only after external audits, meaning the practical payoff for users arrives in stages rather than at one switch. What would contradict that read is an audit finding in the aggregation logic that forces a redesign rather than a patch.

## What to watch

- Whether the next Tachyon components land with published proving-time and bandwidth measurements rather than target figures.
- Whether an external audit of the aggregation logic is commissioned and published before any activation proposal.
- Whether other Zcash wallet and node teams adopt Zakura Common, which determines if the work becomes an ecosystem standard or a single-client feature.