CryptoSlate
A Sept. 16 snapshot of clientdiversity.org displayed three incompatible estimates of Ethereum's largest consensus client at the same time: Blockprint put Teku at 99.83%, Miga Labs put Lighthouse at 51.32%, and Rated put Teku at 53.86%. Each reading comes from a different proxy, and Sigma Prime says the Blockprint classifier behind the 99.83% figure is defunct after the Electra upgrade, even though the dashboard still labels it as updated daily.
Why it matters
Client diversity is a safety property. A bug in a consensus client run by more than 33% of nodes could halt finality, and a critical bug in a client holding a two-thirds majority could finalize an incorrect chain, exposing validators to slashing or costly exits. None of the current estimates answers the question researchers actually care about: the distribution of stake, not the count of visible machines, across clients.
The measurement problem may get harder. Vitalik Buterin's Lean-chain privacy proposal would rotate validator keys daily and hide deposit and withdrawal links, eroding the persistent identifiers that tools like Rated use to group validators by operator. Researchers are racing to build stake-authenticated, privately aggregated client reporting before those traces disappear.
Market impact
Until credible, authenticated reporting exists, the network's headline safety metric is effectively unverifiable, which raises the stakes for any client-clustered bug. A 2025 USENIX study found four observer nodes could locate over 15% of validators, showing both the power and the privacy cost of current tracing methods. Watch for progress on Nethermind-style private aggregation research and any dashboard standards that publish uncertainty alongside estimates.
Frequently asked questions
-
Why did clientdiversity.org show three different Ethereum client shares at once?
The site aggregates independent estimates that each measure a different signal: Blockprint classifies block behavior with machine learning, Miga Labs crawls peer metadata, and Rated uses an undisclosed method. On Sept. 16 these produced 99.83% Teku, 51.32% Lighthouse, and 53.86% Teku respectively.
-
Why is the Blockprint Teku estimate of 99.83% considered unreliable?
Sigma Prime's archived repository says the classifier is no longer accurate after Ethereum's Electra upgrade and considers the project defunct. Clientdiversity.org nevertheless still labels the Blockprint panel as updated daily.
-
What happens if one Ethereum consensus client crosses 33% or 66% of the network?
A bug in a client used by more than 33% of nodes could prevent finality, a liveness failure. A critical bug in a client with a two-thirds majority could finalize an incorrect chain, a safety failure that could expose validators to slashing or forced exit-and-re-entry.
-
How would Ethereum's Lean privacy proposal change client diversity measurement?
The proposal would rebuild the active validator registry daily with fresh keys and use hiding commitments for deposits and withdrawals. That would break the persistent identifiers, like deposit-address grouping, that analysts use to track operator and stake concentration over time.
-
What is the proposed replacement for current client share tracking?
Researchers, including a Nethermind project, are exploring private aggregate reporting where validators encrypt their client choices and prove ballots valid cryptographically, revealing only the aggregate. Open questions remain around authentication, sampling, fake data, and separately measuring operator concentration.
Lookonchain
WuBlockchain
CoinDesk
Crypto News
TheBlock