# Client Diversity of Ethereum Source: https://clientdiversity.org Kiwi News: https://news.kiwistand.com/stories/Client-Diversity-of-Ethereum?index=0x65b6cc7af21b1afab061a11d26e1c779a77d90443f6833a5283585ddd839bc3c97e9cdb4 Submitted by tim on 2024-01-28 UTC, 8 upvotes, 0 comments ## Summary Client diversity is presented as critical to Ethereum's resilience, since a single client used by 66% of validators could finalize a buggy fork, leaving validators who exit it open to a full 32 ETH slashing penalty. Clients above 33% share could prevent finality, so the stated goal is under 33% for each client, with danger beginning above 66%. Multinode validator clients such as Vero and Vouch are recommended because they require M-of-N connected clients to agree before voting, reducing correlated failures from client bugs. Supermajority.info reports 59.4% self-reported network coverage, with the remainder assumed to be mostly Geth. ## Article Diversify Now Improve Ethereum's resilience by using a minority client Switch Clients Dashboard Learn More Client Distribution View Staking Pool Diversity Goal: <33% | Danger: >66% Execution Clients The Ethernodes data has been removed. The data is a representation of what clients are showing free peers and it's not well understood how this correlates to an overall picture of what clients are currently in use. For now, survey data (from execution-diversity.info) is considered the most reliable source of information for the state of execution client diversity. Read more Data provided by Ethernodes — updated daily. Data may not be 100% accurate. (Read more) This data set is mostly comprised of mostly time-intensive manually-gathered data which requires reaching out to each entity. Most of this data hasn't been updated in over a year. Lido curated operator set last updated for Q4 2025 data. This data set does not take into account consolidations so it skews results towards older data. Coinbase data last updated for Q1 2025 data. This data set does not take into account consolidations so it skews results towards older data. Graffiti data is updated on an ongoing basis, but accounts for a small portion of the data set. Data provided by supermajority.info — updated manually. Data may not be 100% accurate. (Read more) Based on 59.4% self-reported network coverage with the remaining assumed to be mostly Geth. Client Diversity Is Not Optional Many know client diversity is important for a more resilient network, but they don't understand why or just how essential it is. It's not only important — it's critical. If a single client is used by 2/3rds (66%) of validators, there's a very real risk this can result in disrupting the chain and monetary loss [1, 2] for node operators. It takes 2/3rds of validators to reach finality. If a client with 66%+ of marketshare has a bug and forks to its own chain, it'll be capable of finalizing. Once the fork finalizes, the validators cannot return to the real chain without being slashed. If 66% of the chain gets slashed simultaneously, the penalty is the whole 32 ETH. So why is >33% marketshare still undesirable? If a client with >33% marketshare forks, it will prevent the chain from finalizing. That's why <33% marketshare is the goal for all clients. Execution clients are not immune. The risks mentioned above apply to both consensus clients and execution clients equally. Client Resources Consensus Clients Client Github Docs Chat Status Support Language Donate Caplin (w/Erigon) stableLinux, Win, macOS, ARMGolang Grandine stableLinux, macOS, ARMRust Lambda deprecatedLinux, macOS, WinElixir- Lighthouse stableLinux, Win, macOS, ARMRust Lodestar stableLinux, Win, macOSTypeScript Nimbus stableLinux, Win, macOS, ARMNim- Prysm stableLinux, Win, macOS, ARMGolang Teku stableLinux, Win, macOSJava Note: Donations made to Protocol Guild are distributed among Ethereum protocol contributors, including client teams. All recipients and splits can be seen here. Execution Clients Client Github Docs Chat Status Support Language Donate Akula deprecated-- Besu stableLinux, Win, macOSJava Erigon stableLinux, Win, macOS, ARMGolang EthereumJS deprecatedLinux, Win, macOSTypeScript- Ethrex stableLinux, macOS, ARMRust- Geth stableLinux, Win, macOS, ARMGolang Nethermind stableLinux, Win, macOS, ARM.NET Nimbus (w/Nimbus CL) alpha-Nim- Reth stableLinux, Win, macOS, ARMRust- Silkworm -pre-alphaLinux, Win, macOSC++ Note: Donations made to Protocol Guild are distributed among Ethereum protocol contributors, including client teams. All recipients and splits can be seen here. Lean Consensus Clients Multinode Clients Feature / Property Vero Vouch Attestation Strategies M of N M of N, First, Best Signer Web3Signer Dirk Distributed Validator Keys No, uses whole key Yes, uses sharded keys Active-active Redundancy Vero Sponsors only Yes Ethereum Remote Signing API Yes No Slashing Protection Yes Yes Slashing Detection Yes No Ease of Migration Designed for simple adoption and rollback More complex setup and integration Eth Docker Support Yes Yes Open Source Yes Yes Dependency Surface Minimal codebase and external dependencies; single maintainer Large/complex codebase (~5x Vero) and external dependencies; creator retired, codebase maintained by Bitwise Typical Use Case Simple way to protect validators from client consensus bugs Highly-configurable multi-node validator setups with sharded validator keys Switch Clients Multinode Clients (Recommended) Multinode validator clients provide protection against client bugs that can cause validators to vote for an invalid fork of the chain. They do so by combining data from multiple consensus and execution clients, requiring M-of-N connected clients to agree before casting their vote. When one client diverges or behaves inconsistently, the validator client continues operating through the rest of the clients as long as M of them agree on which fork to vote for. This way a defect or downtime in any single client does not disrupt validation activity. This approach protects both the operator and the Ethereum network by reducing correlated failures, lowering risks during consensus bugs and improving overall resilience. See the Multinode Clients section to understand if Vero or Vouch is better for you. Under most situations Vero will cover 99% of usecases including institutional staking operations. Standalone Clients For an automated tool (with a GUI) to switch execution clients, see Accidental-Green's Ethereum Client Switcher Submit Guide Error: Select both To and From clients. From Client To Client There are no guides for this migration yet. Submit Guide Resources