DeFi Dispatch is P2P.org's twice-monthly roundup of DeFi developments for institutional participants navigating the intersection of traditional and on-chain finance. Each edition covers the signals that matter for asset managers, custodians, hedge funds, ETF issuers, exchanges, and staking teams operating at the frontier of institutional DeFi and proof-of-stake infrastructure.
Missed the previous edition? Catch up here: DeFi Dispatch: DeFi News and Signals August 2026 (Issue 2)
Short on time? Here are the key takeaways. For the full analysis, continue reading below.
The first half of September brought five developments that institutional participants in DeFi and staking infrastructure should track closely.
The first half of September 2026 is defined by a tension between strong institutional demand signals and a technical upgrade that is taking longer to finalize than developers planned. ETH ETF inflows are positive into September, the validator deposit queue remains deep, and new institutional staking products are launching on Lido V3's modular infrastructure. At the same time, Glamsterdam's finalization problems on Devnet-9 have pushed the October 6 Sepolia target into question and opened a realistic path toward a 2027 mainnet. For institutions building staking programs, the infrastructure is maturing faster than the upgrade timeline, and the demand environment remains structurally constructive regardless of when Glamsterdam activates.
Below, we break down five key developments and why they matter for asset managers, custodians, hedge funds, ETF issuers, exchanges, and staking teams.
U.S. spot Ethereum ETFs recorded approximately $127.7 million in cumulative net inflows through the first four completed sessions of September, according to Farside Investors data, following the $1.42 billion ten-session streak that defined the second half of August. Daily flows have been uneven: $8.6 million on September 1, $48.2 million in outflows on September 2, $141.4 million on September 3, and $25.9 million on September 4. The net figure is positive but significantly below the August pace, reflecting a market consolidating rather than accelerating after a 27% ETH price rally. BlackRock's ETHA and ETHB together accounted for $218.2 million during the week ending September 4, slightly more than the total category net inflow after outflows from competing products were included.
ETH entered September near $2,440 after rallying from below $1,950 in mid-August. The $2,550 level has rejected two breakout attempts and is identified by multiple analysts as the threshold between continued consolidation and a path toward $2,800. Institutional spot buying rather than derivatives expansion has characterized the recovery, making the rally structurally more durable than leverage-driven moves but also more dependent on sustained ETF inflow momentum to clear the $2,550 resistance.
Source: Bitcoin Ethereum News, CryptoTimes, September 2026.
Glamsterdam-Devnet-9 launched on September 1 with 1,000 validators, the largest devnet by node count in the upgrade's testing history, but failed to achieve consistent finalization as of early September. The finalization failure was traced to two issues. First, Devnet-8 exposed a consensus-layer bug where creating a block with the same hash as its parent could cause the entire network to stop. Second, EIP-8037, which introduces a state gas dimension adding a separate gas cost to transactions that create new accounts, contracts, or storage entries, required implementation updates from every execution layer client. Developers agreed at the September 3 All Core Developers Consensus call to target October 6 at 13:53 UTC for the Glamsterdam fork on the Sepolia public testnet, at epoch 351232, while acknowledging explicitly that the date is conditional on Devnet-11 achieving stable finalization first.
Devnet-11 is scheduled to launch on September 14 with 84,000 validators, its Gloas fork planned two days later. Protocol specialist Christine D. Kim noted that all eyes are on Devnet-11, with the October 6 Sepolia date more likely to slip again than to hold if finalization problems persist. Ethereum Foundation engineer Parithosh Jayanthi cautioned that if the next devnet fails, developer calls would weigh skipping the October 6 fork altogether. December 2026 is discussed informally as a possible mainnet window. A further slip into 2027 remains explicitly on the table.
Source: CryptoPotato, 247 Wall St, COINOTAG, September 2026.
Frgmnt, a stablecoin yield protocol on Base, announced on September 11 that its fUSD stablecoin and yield-bearing sfUSD infrastructure are now accessible through institutional custody rails for the first time, marking one of the first examples of a regulated custody environment wrapping DeFi yield mechanics inside a compliance-friendly structure. fUSD is minted on a 1:1 basis against USDC and deployed across selected on-chain lending markets. Holders who stake fUSD receive sfUSD, which earns rewards tied to those lending market conditions. As of September 4, sfUSD was generating 13.32% in protocol staking rewards from on-chain lending markets, though returns shift as underlying lending market conditions evolve.
The protocol operates under a capped, invite-only beta with approximately $100,000 in total value locked according to DeFiLlama data, reflecting an early-stage design that controls collateralization ratios through capped deposit waves rather than open-ended inflows. The next deposit wave is scheduled to open September 15, 2026. The protocol has undergone a CertiK audit. The integration removes the need for institutions to establish separate custody arrangements to use the protocol's products, giving funds, corporate treasuries, and fintechs a more direct route to its stablecoin tools.
Source: Bitcoin.com News, Gokhshtein, September 2026.
Lido V3's stVaults modular infrastructure added new institutional and public vault products on September 9, expanding the ecosystem of configurable Ethereum staking arrangements available to asset managers, ETF issuers, DAOs, and enterprise treasuries since Lido V3's January 30, 2026 mainnet. The institutional vault product offers configurable, non-custodial staking with client-controlled parameters covering custody, permissions, fees, and liquidity. Each institutional vault is isolated, with assets segregated and accounting maintained per vault. The public vault combines ETH staking with EarnETH, Lido's ETH DeFi strategy, with participants receiving an ERC-20 token representing their vault share while the DeFi Wrapper automatically mints stETH and deploys it into EarnETH across a curated set of established DeFi protocols.
The stVaults smart contracts have undergone audits by Certora, including formal verification, MixBytes, Consensys Diligence, Composable Security, Ackee Blockchain, and Sigma Prime, with an ongoing Immunefi bug bounty offering up to $2 million in rewards. Lido's 2026 goals for stVaults include staking 1 million ETH through custom vaults, representing approximately $3 to $4 billion at current prices.
Source: Metaverse Post, Blockchain Reporter, September 2026.
The Ethereum validator deposit queue stood at approximately 1.95 million ETH with an estimated processing time of 34 days as of September 8, 2026, according to beaconcha.in data. The queue is smaller than the 2.23 million ETH recorded in mid-August but remains structurally significant, reflecting sustained demand for new validator entries even after the August ETF-driven inflow surge. Layer 2 networks processed approximately 25.01 million daily transactions as of the September 7 UTC session, compared with 1.86 million on Ethereum's base layer, confirming the scale of activity that Glamsterdam's gas-limit expansion is designed to support when it eventually activates.
The deposit queue, ETF inflow data, and exchange supply dynamics together describe a consistent picture: more ETH is being committed to long-term staking and custody arrangements than is being made available for immediate market transactions. Over one-third of all circulating ETH remains staked, exchange balances are at multi-year lows, and the deposit queue shows no sign of normalizing toward pre-2026 levels.
Source: CryptoTimes, Yahoo Finance, September 2026.
The first half of September 2026 surfaces five converging signals for institutional participants in on-chain infrastructure:
A 1.95 million ETH deposit queue with a 34-day processing time confirms that institutional demand for new validator entries remains elevated well beyond the August ETF inflow surge. For institutions planning new staking program entries, the queue means capital committed today will not be earning staking rewards for over a month. This lag should be factored into return calculations and operational timelines for any staking program launched in Q4 2026.
Yes, but the update is about timeline uncertainty rather than abandoning preparation. EIP-8037's state gas dimension requires infrastructure updates across custody systems, execution venues, and reporting tools regardless of when mainnet activates. The operational risk of waiting for a confirmed mainnet date before beginning preparation is that the preparation window compresses if Devnet-11 succeeds and testing moves quickly. Treat preparation as ongoing rather than event-driven.
Ethereum consensus staking yield of approximately 2.6% to 3.0% comes from protocol issuance and transaction fees earned by validators securing the network. It is denominated in ETH and reflects the economics of proof-of-stake network participation with well-understood risk parameters. sfUSD protocol staking rewards, which were 13.32% as of September 4, come from deploying USDC-backed collateral into on-chain lending markets. They are denominated in stablecoin terms and reflect lending market conditions, credit risk, and liquidity constraints specific to those protocols. The two yield sources serve different institutional purposes and carry fundamentally different risk profiles.
About P2P.org
Founded in 2018, P2P.org helps institutional capital protect Digital Asset Yield across non-custodial staking infrastructure and curated DeFi strategies. With over $10B in assets secured and operating on 35+ proof-of-stake networks, P2P.org maintains a zero-slashing-incident track record, is trusted by over 190 institutional clients and is SOC 2 Type II attested and ISO/IEC 27001:2022 certified. To explore how P2P.org can support your institution's staking or DeFi infrastructure needs, get in touch with our team.
Disclaimer
This material is provided for informational purposes only and does not constitute investment, financial, legal, or tax advice. P2P.org accepts no liability for any actions taken based on it. Latency and performance figures referenced are estimates based on internal benchmarks and may vary depending on network conditions, geography, and client infrastructure. Past performance is not indicative of future results.
<p>Each month, we publish a full breakdown of how our Solana validators performed against the rest of the field: gross rewards, MEV capture, and reliability, all sourced from our own on-chain data collectors and reproducible from raw epoch data. Here's how we performed in August, covering epochs 1010 to 1025. </p><p><strong>The headline</strong></p><p>Almost every Solana validator earns close to the same base reward. Issuance, the protocol-set portion of staking rewards, is identical for any correctly run validator. What actually separates operators is what they capture on top of that base: MEV and reliability.</p><p>In August, P2P.org's Total Gross APY came in at 6.16%, just ahead of the next-best validator we track at 6.11%. The gap is small because most of that return is the same base reward every validator earns. </p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2026/09/1600x900--47-.png" class="kg-image" alt="Bar chart comparing P2P.org's 6.16% Total Gross APY against the next-best peer validator's 6.11% for August 2026." loading="lazy" width="1600" height="900" srcset="https://p2p.org/economy/content/images/size/w600/2026/09/1600x900--47-.png 600w, https://p2p.org/economy/content/images/size/w1000/2026/09/1600x900--47-.png 1000w, https://p2p.org/economy/content/images/2026/09/1600x900--47-.png 1600w" sizes="(min-width: 720px) 720px"></figure><p><strong>Where the edge comes from</strong></p><p>MEV capture: P2P.org's Jito tips rate ran at 0.39%, against a 0.26% simple network average. That's the one part of the reward validators actually compete on: tip volume carries a real element of market luck, but capturing it consistently comes down to validator-client configuration, MEV-strategy setup, and being present for every assigned slot.</p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2026/09/1600x900--44--2.png" class="kg-image" alt="Bar chart comparing P2P.org's 0.39% Jito tips APY against a 0.26% simple network average for August 2026." loading="lazy" width="1600" height="900" srcset="https://p2p.org/economy/content/images/size/w600/2026/09/1600x900--44--2.png 600w, https://p2p.org/economy/content/images/size/w1000/2026/09/1600x900--44--2.png 1000w, https://p2p.org/economy/content/images/2026/09/1600x900--44--2.png 1600w" sizes="(min-width: 720px) 720px"></figure><p>Reliability: Across vote success, block production, and uptime, P2P.org ran ahead of the network average for validators with 100k+ SOL staked, on every metric, in August:</p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2026/09/1600x900--45-.png" class="kg-image" alt="Table comparing P2P.org against network average across three reliability metrics for August 2026: vote success at 99.8% versus 99.0%, block production at 100.0% versus 99.5%, and uptime at 100.0% versus 99.4%." loading="lazy" width="1600" height="900" srcset="https://p2p.org/economy/content/images/size/w600/2026/09/1600x900--45-.png 600w, https://p2p.org/economy/content/images/size/w1000/2026/09/1600x900--45-.png 1000w, https://p2p.org/economy/content/images/2026/09/1600x900--45-.png 1600w" sizes="(min-width: 720px) 720px"></figure><p><strong>Why this matters going forward</strong></p><p>Solana's SGP-0002 vote, passed in August, doubles the pace at which the issuance (base reward) component declines over the next three years. As that shared portion shrinks, MEV and reliability make up more of what a validator actually earns, and more of what separates one from another.</p><p>The full breakdown, including the reward-composition chart, the full peer comparison, and our methodology, is available here: </p><h3 id="read-the-august-2026-solana-staking-snapshot"><a href="https://2e4kdb.share-eu1.hsforms.com/2vrb-vR6VRtCUQrjQA6mhjg?ref=p2p.org"><strong>Read the August 2026 Solana Staking Snapshot</strong></a><br></h3><p>If you're staking on Solana, or considering it, reach out to your account manager or visit p2p.org to get started.<br></p><div class="kg-card kg-toggle-card" data-kg-toggle-state="close"> <div class="kg-toggle-heading"> <h4 class="kg-toggle-heading-text"><span style="white-space: pre-wrap;">FAQ</span></h4> <button class="kg-toggle-card-icon" aria-label="Expand toggle to read content"> <svg id="Regular" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"> <path class="cls-1" d="M23.25,7.311,12.53,18.03a.749.749,0,0,1-1.06,0L.75,7.311"></path> </svg> </button> </div> <div class="kg-toggle-content"><p dir="ltr"><b><strong style="white-space: pre-wrap;">What is Total Gross APY?</strong></b><span style="white-space: pre-wrap;"> It's the combined annualized rate from all three Solana validator reward types: staking (issuance), Jito tips, and block rewards, before any commission is deducted.</span></p><p dir="ltr"><b><strong style="white-space: pre-wrap;">Why does P2P.org report a different APY than sites like Staking Rewards?</strong></b><span style="white-space: pre-wrap;"> A few reasons: whether compounding is included (APY vs. APR), whether commission is deducted (gross vs. net), which reward types are counted, and how the averaging window is defined. None of these methodologies is wrong; they're just measuring different things. We publish our full methodology alongside the raw data on our dashboard.</span></p><p dir="ltr"><b><strong style="white-space: pre-wrap;">Is this the APY I'd actually earn if I staked with P2P.org?</strong></b><span style="white-space: pre-wrap;"> This report shows gross rewards, before commission. Your actual net rewards depend on P2P.org's fee and your specific delegation. Contact your account manager for the exact numbers.</span></p><p dir="ltr"><b><strong style="white-space: pre-wrap;">How often is this published? </strong></b><span style="white-space: pre-wrap;">Monthly, covering the prior month's epochs.</span></p><p dir="ltr"><b><strong style="white-space: pre-wrap;">Can I check these numbers myself?</strong></b><span style="white-space: pre-wrap;"> Yes. All the raw epoch-level data behind this report is exportable from our public dashboard at </span><a href="http://reports.p2p.org/superset/dashboard/p/WMGBkJ8LvPz/?ref=p2p.org"><span style="white-space: pre-wrap;">reports.p2p.org</span></a><span style="white-space: pre-wrap;">, along with the calculation methodology.</span></p></div> </div><p><strong>Disclaimer</strong></p><p><em>This material is provided for informational purposes only and does not constitute investment, financial, legal, or tax advice. </em><a href="http://p2p.org/?ref=p2p.org"><em>P2P.org</em></a><em> accepts no liability for any actions taken based on it. Latency and performance figures referenced are estimates based on internal benchmarks and may vary depending on network conditions, geography, and client infrastructure. Past performance is not indicative of future results.</em></p>
from p2p validator
<h2 id="series-validator-playbook-mev-infrastructure"><strong>Series: Validator Playbook | MEV Infrastructure</strong></h2><p>The Validator Playbook is <a href="http://p2p.org/?ref=p2p.org">P2P.org</a>'s infrastructure education series for institutional Ethereum operators. Each article addresses a specific operational, risk, or governance decision that validator infrastructure teams, staking product managers, ETF issuers, custodians, asset managers, and risk committees face when building or evaluating proof-of-stake infrastructure.</p><p>Previously in the series: <a href="https://p2p.org/economy/validator-playbook-ethereum-staking-etf-validator-infrastructure/">How ETF Issuers Should Evaluate Validator Infrastructure: An Operational Guide</a></p><hr><h2 id="learnings-for-busy-readers">Learnings for Busy Readers</h2><ul><li>MEV-Boost revenue typically adds 10 to 30% to staking rewards. On a 2.78% base APR, that translates to roughly 0.28 to 0.83% additional yield per validator. For institutional operators running thousands of validators, that difference is material.</li><li>The MEV-Boost architecture splits the role of producing a block into four functions: searchers identify MEV opportunities, builders aggregate them into blocks, relays connect builders to validators, and validators propose the selected block to the chain. Validators sit at the end of that chain, not the beginning.</li><li>Of the seven major MEV-Boost relays, only three do not censor according to OFAC compliance requirements. For regulated institutions, relay selection is a compliance decision, not just a yield decision.</li><li>Running multiple relays simultaneously is standard practice. MEV-Boost compares bids across all connected relays and selects the highest-paying valid block for each slot.</li><li>Approximately 88% of Ethereum blocks are currently assembled off-chain through MEV-Boost relays. Running without MEV-Boost means building blocks locally and forgoing the yield uplift from the builder marketplace entirely.</li><li>Glamsterdam's EIP-7732 will move the proposer-builder separation mechanism into the Ethereum protocol itself, replacing the trust-based relay layer with in-protocol bids, commitments, and payments. The relay stack as it exists today has a defined end date.</li><li>The MEV decision for institutional operators involves three distinct layers: the yield layer (relay selection and multi-relay configuration), the compliance layer (OFAC screening and transaction filtering), and the forward-planning layer (preparing for Glamsterdam's ePBS transition).</li></ul><h2 id="why-mev-now-matters-at-the-institutional-level">Why MEV Now Matters at the Institutional Level</h2><p>Ethereum staking rewards come from two structurally different sources that carry different risk profiles and require different infrastructure decisions.</p><p>Consensus layer rewards are protocol-issued and relatively predictable. Validators earn them by attesting to the state of the chain, proposing blocks, and participating in sync committees. Execution layer rewards come from user priority fees and MEV at the time of block proposals. They are variable and market-driven. In Q1 2026, consensus layer rewards accounted for approximately 93% of total validator rewards, with execution layer rewards comprising the remaining 7%.</p><p>That 7% is not a rounding error for institutional operators. For institutional operators running thousands of validators, the MEV uplift is the difference between a profitable and marginal operation. The variability also matters: execution layer rewards spike during periods of elevated on-chain activity, including token launches, liquidation cascades, and major market events, and compress during quieter periods. Managing that variability through relay configuration is where institutional MEV strategy begins.</p><p>Before Pectra, an institution staking 2,048 ETH ran 64 separate validators, each with its own block proposal slot. Post-consolidation, that same position sits in a single validator. Each block proposal now represents a larger concentrated opportunity to capture execution layer rewards. Validator consolidation has made MEV-per-proposal more consequential, not less.</p><h2 id="how-the-mev-boost-stack-works">How the MEV-Boost Stack Works</h2><p>Understanding the institutional decision requires understanding the architecture. The MEV-Boost proposer-builder separation architecture splits the role of producing a block into three separate functions: searchers identify MEV opportunities and submit bundles of transactions that capture them; builders aggregate searcher bundles and other transactions into proposed blocks; relays connect builders to proposers and provide the trust layer that allows proposers to commit to blocks they have not directly constructed.</p><p>MEV-Boost is a sidecar for the beacon node, a separate piece of open-source software, which queries and outsources block-building to a network of builders. Block builders prepare full blocks, optimizing for MEV extraction and fair distribution of rewards. They then submit their blocks to relays. Relays aggregate blocks from multiple builders to select the block with the highest fees. One instance of MEV-Boost can be configured by a validator to connect to multiple relays.</p><p>For institutional operators, three properties of this architecture are operationally significant.</p><p>First, the relay is a trusted intermediary. Validators commit to a block header before seeing the full block contents. They are trusting the relay not to manipulate or front-run the payload. The trust assumption sits entirely outside Ethereum's protocol rules, which is precisely what Glamsterdam's ePBS addresses.</p><p>Second, relay selection is the primary lever. Validators do not choose builders directly. They choose relays, and relays determine which builders they aggregate. The relay decision shapes both the yield profile and the compliance posture of the validator's block production.</p><p>Third, multi-relay configuration is the standard operating model. If connected to multiple relays, the validator's MEV-Boost software can compare bids and select the block header that will provide the highest value, increasing the overall staking reward for that block. Running a single relay concentrates both yield and operational risk on that relay's availability and performance.</p><h2 id="the-compliance-dimension-of-relay-selection">The Compliance Dimension of Relay Selection</h2><p>For institutional validators operating under regulatory obligations, relay selection carries a compliance dimension that yield calculations alone do not capture.</p><p>Of the seven major MEV-Boost relays, only three do not censor according to OFAC compliance requirements. The remaining relays apply transaction filtering based on OFAC sanctions lists, meaning they will not include transactions involving sanctioned addresses in the blocks they produce. For regulated institutions, using a non-OFAC-compliant relay can expose the validator to sanctions risk if a block they propose contains a transaction involving a sanctioned counterparty.</p><p>The compliance question involves two distinct considerations that institutions need to evaluate separately.</p><p>The first is sanctions exposure. Regulated institutions operating under US jurisdiction, or serving clients subject to US regulatory oversight, typically need to confirm that their block production infrastructure does not facilitate transactions with sanctioned addresses. OFAC-compliant relays address this by filtering their block payloads against the SDN list before forwarding bids to validators.</p><p>The second is censorship posture. OFAC-compliant relays filter transactions, which means they exclude certain transactions from blocks they produce. For institutions with clients or stakeholders concerned about Ethereum's censorship resistance properties, running OFAC-compliant relays exclusively carries a network-level implication beyond the institution's own compliance posture.</p><p>The practical resolution for most institutional operators is a multi-relay configuration that includes both OFAC-compliant and non-censoring relays, with MEV-Boost selecting the highest bid at the time of each block proposal. This approach maintains compliance coverage on most slots while preserving access to the full builder market when a non-censoring relay wins the bid.</p><p>Source: <a href="https://www.kucoin.com/blog/ethereum-staking-in-2026-yield-trends-validator-queue-dynamics-and-mev-impact-exlained?ref=p2p.org">KuCoin</a></p><figure class="kg-card kg-image-card kg-card-hascaption"><img src="https://p2p.org/economy/content/images/2026/09/Diagram-showing-the-MEV-Boost-stack-from-searchers-to-validators-on-the-left--and-the-three-institutional-decision-layers-on-the-right--yield-layer--compliance-layer--and-forward-planning-layer-for-Glamsterdam-ePBS..jpg" class="kg-image" alt="Diagram showing the MEV-Boost stack from searchers to validators on the left, and the three institutional decision layers on the right: yield layer, compliance layer, and forward planning layer for Glamsterdam ePBS." loading="lazy" width="1600" height="900" srcset="https://p2p.org/economy/content/images/size/w600/2026/09/Diagram-showing-the-MEV-Boost-stack-from-searchers-to-validators-on-the-left--and-the-three-institutional-decision-layers-on-the-right--yield-layer--compliance-layer--and-forward-planning-layer-for-Glamsterdam-ePBS..jpg 600w, https://p2p.org/economy/content/images/size/w1000/2026/09/Diagram-showing-the-MEV-Boost-stack-from-searchers-to-validators-on-the-left--and-the-three-institutional-decision-layers-on-the-right--yield-layer--compliance-layer--and-forward-planning-layer-for-Glamsterdam-ePBS..jpg 1000w, https://p2p.org/economy/content/images/2026/09/Diagram-showing-the-MEV-Boost-stack-from-searchers-to-validators-on-the-left--and-the-three-institutional-decision-layers-on-the-right--yield-layer--compliance-layer--and-forward-planning-layer-for-Glamsterdam-ePBS..jpg 1600w" sizes="(min-width: 720px) 720px"><figcaption><i><em class="italic" style="white-space: pre-wrap;">The MEV-Boost architecture mapped against the three institutional decision layers: yield configuration, compliance posture, and Glamsterdam readiness. All three converge at the relay node.</em></i></figcaption></figure><h2 id="what-glamsterdam-changes-for-the-mev-stack">What Glamsterdam Changes for the MEV Stack</h2><p>The MEV-Boost relay architecture is transitional infrastructure. Glamsterdam's enshrined Proposer-Builder Separation, EIP-7732, formally separates the job of the proposer from the builder and enshrines this hand-off directly into the protocol. Building the trustless exchange of a block payload for payment directly into the protocol removes the need for third-party middleware like MEV-Boost.</p><p>Under ePBS, builders cryptographically seal their blocks and commit to a bid. Validators select the highest bid without seeing transaction contents, and the block is only revealed after the commitment is locked in. The trust assumption that currently sits with the relay moves into the protocol itself.</p><p>Glamsterdam has been pushed back from the first half of 2026 to the fourth quarter of 2026. As of September 2026, the upgrade is in active testnet preparation. Glamsterdam is currently in the devnet phase, where client teams test the new rules on small coordinated networks. Public testnets like Sepolia and Hoodi will activate before mainnet.</p><p>For institutional operators, Glamsterdam's ePBS transition has three operational implications worth tracking now rather than at activation.</p><p>The relay dependency changes but does not disappear immediately. Builders and proposers might still choose to use off-protocol relays or middleware for complex features that are not yet part of the core protocol. The current relay infrastructure will likely operate in parallel with the new in-protocol mechanism during a transition period.</p><p>Compliance architecture needs to be reconsidered. Under the current MEV-Boost model, OFAC compliance is implemented at the relay layer. Under ePBS, the filtering logic will need to be remapped to the new in-protocol builder selection mechanism. Institutions that have built compliance programs around relay-level OFAC filtering will need to evaluate how those programs need to adapt.</p><p>Gas limit expansion affects execution layer reward dynamics. Glamsterdam will bring bigger blocks, moving from a gas limit of 60 million per block to a gas limit floor of 200 million per block. Whether that expansion increases execution layer rewards depends on how demand for blockspace responds to the increased supply. Institutions modeling execution layer yield projections for staking programs should treat Glamsterdam-era reward assumptions as uncertain until mainnet data is available.</p><p>Source: <a href="https://ethereum.org/roadmap/glamsterdam/?ref=p2p.org">ethereum.org</a></p><div class="kg-card kg-callout-card kg-callout-card-blue"><div class="kg-callout-emoji">🗞️</div><div class="kg-callout-text"><b><strong style="white-space: pre-wrap;">The institutional digital asset space moves fast.</strong></b><br><br>Our subscribers get structured analysis across staking, DeFi vaults, and regulation through <i><em class="italic" style="white-space: pre-wrap;">DeFi Dispatch</em></i>, <i><em class="italic" style="white-space: pre-wrap;">Institutional Lens</em></i>, <i><em class="italic" style="white-space: pre-wrap;">DeFi Infrastructure for Institutions</em></i>, and <i><em class="italic" style="white-space: pre-wrap;">Legal Layer</em></i>.<br><br>No noise. Just the signals that matter.<br><br><b><strong style="white-space: pre-wrap;">Subscribe to the newsletter at the bottom of this page.</strong></b></div></div><h2 id="the-mev-configuration-checklist-for-institutional-operators">The MEV Configuration Checklist for Institutional Operators</h2><p>The following checklist covers the MEV infrastructure decisions that institutional validators need to have documented and actively managed.</p><h3 id="relay-selection-and-configuration">Relay selection and configuration</h3><p>[ ] MEV-Boost installed and running as a sidecar to the beacon node</p><p>[ ] Minimum bid threshold configured to ensure local block production fallback when no relay bid meets the floor</p><p>[ ] Multiple relays connected to maximize bid competition across slots</p><p>[ ] Relay list reviewed for OFAC compliance posture against the institution's regulatory obligations</p><p>[ ] Relay availability and performance monitored independently of MEV-Boost defaults</p><h3 id="compliance-posture">Compliance posture</h3><p>[ ] OFAC-compliant relays confirmed in the active relay list for all validators operating under US regulatory jurisdiction</p><p>[ ] Transaction filtering policy documented and reviewed by compliance team</p><p>[ ] Censorship posture decision made explicitly and recorded: full OFAC filtering, partial filtering, or multi-relay with non-censoring fallback</p><p>[ ] Relay compliance posture reviewed on a defined schedule, as relay policies change</p><h3 id="operational-resilience">Operational resilience</h3><p>[ ] Local block production fallback confirmed and tested: if all relays fail to return a valid bid, the validator must be able to build locally without downtime</p><p>[ ] Relay response time monitoring in place: slow relay responses can cause missed proposals</p><p>[ ] Client updates for MEV-Boost and beacon node coordinated to avoid version incompatibilities at upgrade boundaries</p><p>[ ] Glamsterdam client readiness on the infrastructure roadmap ahead of testnet activation</p><h3 id="reporting-and-audit-trail">Reporting and audit trail</h3><p>[ ] MEV revenue logged per validator per slot for reward attribution and NAV reconciliation</p><p>[ ] Relay source logged per block proposal for compliance audit trail</p><p>[ ] Execution layer reward variance documented for shareholder reporting and fund-level performance reconciliation</p><h2 id="key-takeaway">Key Takeaway</h2><p>MEV-Boost is now standard infrastructure for institutional Ethereum validators. Approximately 88% of Ethereum blocks are assembled off-chain through MEV-Boost relays. Running without it means forgoing the execution layer yield uplift that the builder marketplace generates on every block proposal.</p><p>For custodians, asset managers, ETF and ETP issuers, exchanges, treasury teams, staking product managers, and risk committees, the MEV decision is not a single infrastructure choice. It is three parallel decisions: relay selection for yield maximization, relay compliance posture for regulatory alignment, and forward infrastructure planning for Glamsterdam's ePBS transition.</p><p>The institutions best positioned across all three have explicitly documented their relay configuration, aligned it with their compliance team's review of OFAC filtering requirements, and placed Glamsterdam client readiness on their infrastructure roadmap before the testnet activation schedule forces the conversation.</p><p>P2P.org's validator infrastructure runs MEV-Boost with OFAC-compliant relay configuration as standard across all client accounts. For institutional operators evaluating their MEV infrastructure, <a href="https://p2p.org/contact?ref=p2p.org">get in touch with our team</a>.</p><h2 id="frequently-asked-questions-faq">Frequently Asked Questions (FAQ)<br></h2><h3 id="what-is-mev-and-why-does-it-matter-for-institutional-ethereum-validators">What is MEV and why does it matter for institutional Ethereum validators?</h3><p>MEV stands for Maximal Extractable Value. It refers to the additional value that can be captured during block production by reordering, inserting, or excluding transactions within a block beyond the standard consensus layer rewards. On Ethereum, MEV flows primarily through the block builder marketplace: searchers identify profitable transaction ordering opportunities, builders incorporate them into blocks, and validators capture the value when they propose those blocks. In Q1 2026, execution layer rewards, including MEV accounted for approximately 7% of total validator rewards, with the consensus layer providing the remaining 93%. For institutional operators managing large validator sets, that 7% is a material component of total staking economics and needs to be actively managed rather than left to default configuration.</p><h3 id="what-is-mev-boost-and-how-does-it-work">What is MEV-Boost and how does it work?</h3><p>MEV-Boost is a sidecar for the beacon node, a separate piece of open source software which queries and outsources block-building to a network of builders. Block builders prepare full blocks optimizing for MEV extraction and fair distribution of rewards. They then submit their blocks to relays. Relays aggregate blocks from multiple builders in order to select the block with the highest fees. One instance of MEV-Boost can be configured by a validator to connect to multiple relays. The consensus layer client of a validator proposes the most profitable block received from MEV-Boost to the Ethereum network for attestation and block inclusion. Running MEV-Boost is effectively standard practice for institutional validators because the yield differential between MEV-Boost-enabled and local block production is consistently positive over time.</p><h3 id="why-is-relay-selection-a-compliance-decision-for-regulated-institutions">Why is relay selection a compliance decision for regulated institutions?</h3><p>Not all relays treat transactions equally. Of the seven major MEV-Boost relays, only three do not censor according to OFAC compliance requirements. Relays that apply OFAC filtering exclude transactions involving sanctioned addresses from the blocks they produce. Regulated institutions operating under US jurisdiction, or serving clients subject to US regulatory oversight, need to confirm that their block production infrastructure does not facilitate transactions with sanctioned counterparties. Running a non-OFAC-compliant relay without explicit compliance team sign-off introduces sanctions exposure that cannot be managed after the fact, since the block is already on-chain by the time a sanctions issue would be detected.</p><h3 id="what-is-the-risk-of-running-only-one-relay">What is the risk of running only one relay?</h3><p>Relay concentration introduces two distinct risks. The first is yield concentration: if a single relay consistently loses bid competition against others, the validator systematically underperforms on execution layer rewards relative to a multi-relay configuration. The second is operational concentration: if the single relay experiences an outage, latency issue, or returns invalid bids, the validator either misses proposals or falls back to local block production without the execution layer uplift. Multi-relay configuration addresses both risks by distributing bid competition across multiple providers and providing operational redundancy if any individual relay fails.</p><h3 id="what-does-glamsterdams-eip-7732-mean-for-the-current-mev-relay-stack">What does Glamsterdam's EIP-7732 mean for the current MEV relay stack?</h3><p>Glamsterdam's EIP-7732 moves the proposer-builder separation mechanism into the Ethereum protocol itself. Under ePBS, builders cryptographically seal their blocks and commit to a bid. Validators select the highest bid without seeing transaction contents, and the block is only revealed after the commitment is locked in. Building this trustless exchange directly into the protocol removes the need for third-party middleware like MEV-Boost. However, builders and proposers might still choose to use off-protocol relays or middleware for complex features not yet part of the core protocol. Glamsterdam has been pushed back to Q4 2026 and is currently in active testnet preparation. Institutions should treat current MEV-Boost relay configurations as operational through at least the first half of the 2026 mainnet transition period, while beginning to assess how their compliance architecture will need to adapt to the new in-protocol builder selection mechanism.</p><h3 id="how-should-etf-issuers-think-about-mev-revenue-in-fund-reporting">How should ETF issuers think about MEV revenue in fund reporting?</h3><p>MEV revenue flows through the execution layer as part of block proposal rewards. For ETF products that distribute protocol-attributed participation rewards to shareholders, execution layer rewards, including MEV need to be included in the reward attribution methodology. The variability of MEV revenue complicates distribution modeling: execution layer rewards spike during high-activity periods and compress during quieter ones. ETF issuers building NAV calculation infrastructure for staking-enabled products should model execution layer reward variability as a range rather than a fixed assumption, and confirm with their fund administrator that the reporting infrastructure captures relay-source revenue per block proposal for audit trail purposes.</p><hr><p><strong>About P2P.org</strong></p><p>Founded in 2018, P2P.org helps institutional capital protect digital asset yield across non-custodial staking infrastructure and curated DeFi strategies. With over $10B in assets secured and operating on 35+ proof-of-stake networks, P2P.org maintains a zero-slashing-incident track record, is trusted by over 190 institutional clients and is SOC 2 Type II attested and ISO/IEC 27001:2022 certified. To explore how P2P.org can support your institution's staking or DeFi infrastructure needs, <a href="https://p2p.org/contact?ref=p2p.org">get in touch with our team</a>.</p><p><strong>Disclaimer</strong></p><p>This material is provided for informational purposes only and does not constitute investment, financial, legal, or tax advice. <a href="http://p2p.org/?ref=p2p.org">P2P.org</a> accepts no liability for any actions taken based on it. Latency and performance figures referenced are estimates based on internal benchmarks and may vary depending on network conditions, geography, and client infrastructure. Past performance is not indicative of future results.</p>
from p2p validator