Helsinki, Finland, May 22, 2024 (GLOBE NEWSWIRE) -- Gevulot, a purpose-built L1, is launching the first shared prover layer optimized exclusively for zero-knowledge (ZK) proof generation and verification, starting with its mainnet prover network, Firestarter, going live on July 15. Whitelisted projects will receive public good access to the mainnet. Gevulot also provides an in-depth explanation of ZK Endgame and its approach to ZK, to ultimately bring the compute and cost overhead of verifiability to near zero.
The Gevulot Devnet is currently a free public good testnet for registered users and Galxe community participation is encouraged. Interested developers can start offloading provable compute or start benchmarking proof systems to enjoy production-ready proof generation to offset their existing overhead costs. As Gevulot scales, organizations interested in running prover nodes can apply here for the ZkCloud compute network, which currently has a pipeline of over 10,000 CPUs and GPUs which meet institutional requirements.
Gevulot’s Approach to ZK Endgame
The Gevulot team believes zero-knowledge proving will characterize distributed systems in the near future. Zero-knowledge proofs enhance a blockchain network’s scalability, compatibility with other networks and applications, and privacy. Their purpose is validating computations and improving transaction integrity without revealing additional information.
A growing niche in the industry, including Gevulot, believes today’s infrastructure is not ready to embrace ZK proving yet. On the contrary, most applications rely on permissioned proving from a single prover, drifting away from the fundamental definition of a decentralized ecosystem, which should embody credible neutrality and censorship resistance.
Gevulot aims to solve the issues of present-day proving infrastructures and accelerate zero-knowledge proving adoption by building the ZkCloud, a decentralized compute network optimized for offloading arbitrary provable compute. This solution is slated to help unlock the near limitless compute capacity on blockchains via decentralized zero-knowledge proofs, which would represent the apex of distributed systems evolution. In other words, it would be the zero-knowledge endgame and, as the team puts it on their blog, “nothing short of all the world's compute being verifiable or as close to that as is allowed by physics.”
Zero-knowledge proving is rapidly evolving, drastically reducing the cost of verifying the accuracy of a third party’s computation. This solves the “cost of verifiability” dilemma, which has bothered industry experts since the dawn of distributed systems technology. It also means we can rely on the performance and redundancy characteristics of trusted computation without requiring trust.
The ZK Endgame aims for the compute and cost overhead of verifiability to approach zero. In other words, to reach its goal, zero-knowledge proving must also bring the required compute time to zero, and not only its financial costs.
Reaching absolute zero costs of verifiability is most likely unachievable. However, getting as close as possible to zero for verifiability and compute time may be possible by getting ZK proving on a trend line that involves several new breakthroughs in cryptography, hardware, and other essential metrics, such as bandwidth and/or electricity costs.
Gevulot is building ZkCloud to take zero-knowledge, proving on a trend line to the ZK Endgame while pushing technological frontiers. The project considers itself a contributor to reaching the pinnacle of zero-knowledge proving in distributed systems. As Teemu Päivinen, CEO and Co-founder of Gevulot, describes it:
“In our vision, we are the compute portion of this future cloud and world computer. We believe other teams will provide the data storage and other aspects of this cloud. I don't think it's going to be a single monolithic service. The only realistic chance of getting there is to utilize zero-knowledge proofs. Our short-term goal, which we're very much executing on currently and very excited about, is just lowering the cost considerably and then letting the cryptographers work their magic to make things faster."
ZkCloud is redefining cloud infrastructure with a product that is decentralized, censorship-resistant, and permissionless while providing universal access for proving.
About Gevulot
Gevulot offers the first shared prover layer for the modular stack, with internet-scale ZkCloud compute optimized for zero-knowledge proofs. It is designed to aggregate proving workloads from across the blockchain industry, improving economics for all participants and unlocking unprecedented use cases. Behind Gevulot is a veteran team of builders led by the founder of Equilibrium Labs, a group renowned for their protocol level R&D on projects such as Starknet, Solana, Aleo, Avail, and other significant industry projects.
Gevulot's mainnet prover network, Firestarter, will launch as a public good on July 15. The project has a growing list of prestigious partners, including Espresso, Gelato, Radius, Hylé, 0G labs, AltLayer, ZKM, Snapchain, Horizen Labs, Karnot, Airchains, Zeeve, and over 50 more to be announced. Their mission is to contribute to the ZK Endgame. Interested teams can contact Gevulot now for a potential grant.
Moreover, Gevulot enjoys the support of recognized VC backers with the project raising $6 million in a seed round in mid 2023. They include lead investor Variant and other industry-leading brands, such as Supranational, P2P.org, Staking Facilities, RockawayX Infrastructure, Marc Boiron (Polygon Labs), Shumo Chu (founder of Manta Network & NEBRA), Calvin Liu (CSO of Eigenlayer), George Lamberth (EIR dao5), and Mo Dong (founder of Celer & Brevis Network). Additional funding is still to be announced.
Other important milestones in Gevulot's roadmap include the release of an incentivized compute network in Q3 of 2024 and the ZkCloud mainnet launch, which should occur by 2025's first financial quarter.
You can follow the project’s development and stay up to date at these links: Website | X (Twitter) | Telegram | GitHub | LinkedIn | Docs | Galxe Campaign | Discord
