EigenLayer's restaking sector has undergone a severe contraction in 2026. Total value locked peaked near $20 billion in late 2024/early 2025 before declining to approximately $7 billion by mid-2026, according to DefiLlama data. The EIGEN token, which reached an all-time high of $5.65 in December ...
"Restaking tokens need revenue proof, not just a security narrative." — CryptoDaily editorial analysis, June 2026
EigenLayer's restaking sector has undergone a severe contraction in 2026. Total value locked peaked near $20 billion in late 2024/early 2025 before declining to approximately $7 billion by mid-2026, according to DefiLlama data. The EIGEN token, which reached an all-time high of $5.65 in December 2024, traded at $0.19 as of August 2, 2026 — a 96.6% decline. Market capitalization fell from over $4 billion at peak to approximately $138 million.
The April 2026 KelpDAO exploit — a $292 million theft from a liquid restaking protocol built on EigenLayer — triggered $5.4 billion in withdrawals across the restaking sector and created $177 million in bad debt on Aave, according to KuCoin research. One competitor, Karak, has abandoned restaking entirely, rebranding to OpenGDP and pivoting to real-world economic infrastructure. Its TVL collapsed from $700 million to $6.6 million.
The data raises a structural question: whether the 1-2% yield premium restaking offers above base Ethereum staking (approximately 2.8% APR post-Pectra) is sufficient compensation for the additional smart contract, slashing, and composability risks the category introduces.
EigenLayer's TVL trajectory in 2026 reflects a market that overallocated capital to a category with unproven revenue mechanics. According to Fensory, EigenLayer held $8.9 billion in TVL as of March 2026, down from a peak near $19.7 billion. The protocol retained 93.9% market share of all Ethereum restaking activity at that point, with the remaining 5.5% held by Symbiotic and 0.6% by Karak.
The April 2026 KelpDAO exploit accelerated the decline. The $292 million hack triggered approximately $5.4 billion in withdrawals across the restaking sector, according to multiple reports from CoinDesk and KuCoin. Post-exploit TVL fell to approximately $7 billion, though some recovery has occurred since.
The TVL decline was not uniform. Active address counts on EigenLayer fell to between 40 and 240 daily in early August 2026, according to DefiLlama data — a sharp compression that suggests much of the remaining TVL is held by a small number of large depositors rather than broad-based participation.
The liquid restaking protocol layer saw parallel contraction. Ether.fi, the largest LRT protocol, held approximately $5.6 billion in TVL as of mid-2026. Renzo maintained approximately $3 billion, and Puffer Finance held approximately $1.8 billion. KelpDAO, which held approximately $1.6 billion before the exploit, saw its TVL effectively zeroed. According to 99Bitcoins, LRT protocols collectively lost over $1 billion in TVL following the slashing activation in early 2025.
The EIGEN token has been one of the worst-performing major protocol tokens of 2026. It reached an all-time high of $5.65 in December 2024 and traded at $0.19 as of August 2, 2026, according to MetaMask price data. This represents a 96.6% decline from peak.
Market capitalization fell from over $4 billion to approximately $138 million. Circulating supply has expanded to approximately 872 million tokens out of a total supply of 1.67 billion, according to CoinMarketCap data.
The vesting schedule continues to add supply pressure. Investor and early contributor tokens unlock at 4% monthly following a September 2025 cliff, with full vesting completing in September 2027, per token tracking platform data. A June 30, 2026 unlock released 36.82 million EIGEN tokens worth approximately $8.7 million at prevailing prices. With monthly unlocks of approximately 4% of investor allocations continuing through September 2027, the token faces persistent supply-side overhang.
The rebrand from EigenLayer to EigenCloud in mid-2025 was intended to signal the protocol's evolution from a restaking primitive into a full-stack verifiable cloud platform. The product now encompasses EigenDA (data availability), EigenVerify, and EigenCompute. Whether the rebrand has altered investor perception is unclear; the token has continued declining since.
The April 18, 2026 KelpDAO exploit exposed the composability risk inherent in restaking's multi-layered architecture. According to Chainalysis and Halborn security analyses, attackers — linked to North Korea's Lazarus Group — exploited a 1-of-1 verifier configuration in KelpDAO's LayerZero bridge. They compromised RPC nodes feeding data to the single verifier, injected fake cross-chain messages, and minted 116,500 rsETH (approximately 18% of circulating supply).
The stolen rsETH was then used as collateral on Aave V3 to borrow ETH and WETH before funds were routed through Tornado Cash. According to KuCoin research, the exploit created approximately $177 million in bad debt on Aave's books — a direct demonstration of how restaking composability transmits losses across protocols.
The broader systemic impact was significant. The exploit triggered $5.4 billion in withdrawals across the restaking sector and became the largest DeFi exploit of 2026, according to CoinDesk. Combined with the $285 million Drift Protocol breach on April 1, North Korean groups accounted for 76% of all crypto hack value through April 2026, per TRM Labs data.
The KelpDAO incident was not a smart contract vulnerability in EigenLayer itself. However, it demonstrated that the liquid restaking layer — which issues derivative tokens representing restaked positions — introduces additional attack surface that is difficult to audit at the base protocol level. The 1-of-1 bridge configuration was a deployment choice by KelpDAO, not an EigenLayer design flaw. But the economic damage propagated through EigenLayer's ecosystem regardless.
The most structurally concerning data point for EigenLayer is the gap between TVL and revenue. According to DefiLlama, EigenCloud recorded $71,623 in daily fees in recent data, with $0 captured as protocol revenue. All fees accrued to node operators and stakers, with no value flowing to the protocol or EIGEN token holders.
This creates a fundamental valuation problem. Even at a reduced market capitalization of $138 million, the protocol's annualized fee run rate of approximately $26 million (extrapolating $71K daily) produces no protocol-level revenue. The fees exist, but they flow entirely to operators, not to the protocol treasury or token holders.
For comparison: Ethereum's base layer generated approximately $3.1 billion in annualized fee revenue as of late 2025, according to the webthreepedia foundational economic report. Lido, the largest liquid staking protocol, captures a 10% fee on staking rewards. EigenLayer, as of mid-2026, captures nothing.
Over 20 Actively Validated Services (AVSs) had launched on EigenLayer by mid-2026, with EigenDA processing the highest volume. EigenDA achieved 100 MB/s throughput. However, L2BEAT reported no EigenDA throughput data has been available since May 31, 2026 — a notable gap in public reporting.
The restaking market has consolidated from three competitors to functionally two. Karak, which held $700 million in TVL at peak, rebranded to OpenGDP and exited restaking entirely. Its TVL fell to $6.6 million, according to Protofire's 2026 restaking comparison. The pivot targeted "programmable GDP" and real-world economic infrastructure, with the company's XP staking program winding down.
Symbiotic, which launched as the first fully permissionless restaking protocol, holds approximately 5.5% market share. It differentiates through modular architecture and maximum customization for DeFi-native applications, according to Protofire.
EigenLayer's 94% market share is, paradoxically, both a strength and a systemic risk factor. When one protocol holds nearly all TVL in a category, a systemic event in that protocol propagates across the entire sector. The KelpDAO exploit demonstrated this contagion mechanism. EigenLayer itself was not directly exploited, yet its ecosystem absorbed the bulk of the $5.4 billion in withdrawal pressure.
EigenLayer's governance has recognized the revenue gap. ELIP-12, a proposal launched in Q1 2026, establishes an Incentives Committee to redirect EIGEN emissions toward fee-generating AVSs and introduces two revenue capture mechanisms, according to the EigenFoundation blog.
First, a 20% fee on AVS rewards for subsidized stake — meaning operators who receive EIGEN incentives must share a portion of AVS rewards back to the protocol. Second, 100% of EigenCloud service fees (from EigenDA, EigenCompute, and EigenVerify), after operator costs, flow into a buyback contract. Only fee-paying AVSs remain eligible for incentive allocations.
The proposal represents a strategic shift from "idle stake" toward "productive stake." In theory, it creates deflationary pressure on EIGEN as revenue scales. In practice, the fee base remains small. At $71K daily in total ecosystem fees, a 20% take rate produces approximately $14K per day, or $5.2 million annualized — insufficient to offset monthly token unlocks that add millions in supply pressure.
The success of ELIP-12 depends on whether EigenCloud services, particularly EigenDA, can generate meaningful fee revenue. The gap in EigenDA throughput data since May 2026 makes this assessment difficult.
The restaking value proposition rests on the yield premium above base Ethereum staking. According to KuCoin and multiple staking yield trackers, the current yield landscape is:
| Method | APR Range | Key Risks | |--------|-----------|-----------| | Solo ETH Staking | 3.2% - 3.8% | Consensus-layer slashing only | | Liquid Staking (Lido) | 2.4% - 3.5% | Smart contract risk, 10% fee | | CEX Staking | 2.5% - 3.0% | Custody risk, 25%+ fees | | EigenLayer Restaking | 4.0% - 5.0% | Double slashing, AVS risk, smart contract risk, composability risk | | Peak Restaking (with points) | 10% - 15% | All above + incentive dependency |
The incremental 1-2% yield from restaking must compensate for: (1) double slashing exposure across Ethereum consensus and AVS layers simultaneously; (2) smart contract risk from EigenLayer contracts plus LRT protocol contracts plus AVS contracts; (3) composability contagion, as demonstrated by the KelpDAO exploit's impact on Aave; and (4) liquidity risk during stress events.
The 10-15% yields seen during peak "points farming" periods in 2024-2025 were subsidized by anticipated token distributions, not organic fee revenue. As those incentives normalized, yields compressed to the 4-5% range, reducing the risk-adjusted premium to a level that many institutional allocators have found insufficient.
EigenLayer TVL declined from approximately $20 billion at peak to approximately $7 billion by mid-2026, with daily active addresses falling to 40-240, suggesting concentrated holdings among large depositors.
EIGEN token fell 96.6% from its $5.65 all-time high to $0.19, with market capitalization declining from over $4 billion to $138 million. Monthly 4% vesting unlocks continue through September 2027.
The KelpDAO exploit ($292 million) triggered $5.4 billion in sector-wide withdrawals and created $177 million in bad debt on Aave, demonstrating systemic contagion risk in restaking's composable architecture.
EigenLayer generates approximately $71K in daily ecosystem fees with zero protocol revenue — all fees accrue to operators and stakers, not to the protocol or EIGEN holders.
ELIP-12 introduces a 20% fee on subsidized AVS rewards and directs EigenCloud service fees to a buyback contract, but the current fee base produces only an estimated $5.2 million annualized at the protocol level.
Karak exited restaking entirely (rebranding to OpenGDP; TVL fell from $700M to $6.6M), leaving EigenLayer with 94% market share and Symbiotic with 5.5%.
The restaking yield premium of 1-2% above base Ethereum staking (2.8% APR) introduces double slashing, smart contract, and composability risks that the KelpDAO incident demonstrated are not theoretical.
EigenLayer's restaking experiment represents one of the clearest case studies in 2026 of the gap between capital attraction and revenue generation. The protocol demonstrated it could attract $20 billion in TVL through a combination of yield narrative and points-based incentive structures. It has not yet demonstrated it can generate fee revenue commensurate with that capital base.
The economic fundamentals are stark: $71K in daily fees against billions in TVL produces a fee yield that is negligible. ELIP-12 acknowledges this gap, but the proposed mechanisms — a 20% take rate on subsidized rewards and cloud service fees — are starting from a small base. Whether EigenDA and other EigenCloud products can generate sufficient demand to make these mechanisms meaningful is the central question for the protocol's 2026-2027 trajectory.
The KelpDAO exploit added a real-world data point to the systemic risk discussion. Restaking's composable architecture creates value chains where a vulnerability in one layer (a bridge configuration) can transmit losses across multiple protocols (lending markets). The $177 million in Aave bad debt was not caused by an Aave vulnerability — it was caused by the composability of restaked derivative tokens used as collateral.
The restaking sector is not dead. EigenLayer retains billions in TVL and has introduced governance mechanisms to address its revenue gap. But the data suggests the sector has transitioned from a growth narrative driven by points farming and TVL accumulation to a mature-phase challenge of proving unit economics. At $0.19 per EIGEN — 96.6% below its all-time high — the market has priced in significant skepticism about whether that transition will succeed.