On August 29, the Zakura development team unveiled an open-source suite of cryptographic and protocol libraries for the Zcash ecosystem. Dubbed Zakura Common, the toolset slashes private transaction preparation time in crypto wallets from over three seconds down to under 200 ms in key scenarios.
Client-Side Optimization, Not Network-Level
Crucially, this speedup applies to local client computations performed on the user’s device prior to sending data to the network, rather than blockchain block confirmation times.
These optimizations require no Zcash protocol upgrades or hard forks. Dual-licensed under MIT / Apache 2.0, the libraries are ready for wallet developers to integrate directly into the live mainnet today.
Multi-Fold Performance Gains: Benchmark Results
The team attributed the performance boost to comprehensive component-wide optimizations combining modern cryptographic and compiler techniques with custom Zcash-specific solutions.
Internal benchmark tests by Zakura demonstrated the following improvements:
- Proof generation: Over 14x faster on mobile devices and over 5x faster on desktop computers.
- Sinsemilla hashing: Over 21x faster.
- Trial decryption: Over 1.5x faster.
- zk-SNARK verification: 4x to 8x faster.
New Architecture: Common and Wallet Libraries
The core optimizations reside in the zakura-core/common monorepository, featuring modified and optimized Rust libraries built on librustzcash.
Separately, the team is actively developing zakura-core/wallet-libraries for private wallet integrations and third-party services. Future additions will include Private Information Retrieval (PIR) tools, allowing users to query server data completely anonymously without revealing which specific database item interests them.
While Zakura 1.3.0 was initially announced as an upcoming build, official release v1.3.0 went live on the same day (August 29). The developers migrated their own products to Zakura Common and urged third-party wallets to replace legacy librustzcash builds. Project Tachyon has already adopted the new architecture.
Zcash Scaling and Roadmap Alignment
Zakura nodes already support the 25-second target block interval proposed for NU7 (ZIP 218), down from the current 75 seconds. Even when maintaining equivalent chain-reorganization protection, the actual reduction in latency promises to be substantial.
ZIP 218 also introduces operation limits on Sapling and Orchard private pools. According to calculations, this cuts daily light-wallet sync data by 37% (from 271 MB to 169 MB per day) and reduces trial decryptions from 4.8M to 2.3M.
Orchard’s estimated throughput will more than double, rising from 2.9 to 6.6 transactions per second.
