Ethereum's restaking sector has lost approximately $13 billion in total value locked since its December 2024 peak. EigenCloud, the rebranded EigenLayer platform, holds $5.1 billion in TVL as of August 2026, down 77% from its $22.06 billion high recorded on August 14, 2025. Symbiotic, the nearest ...
"End of an era. Sad. I still think restaking will come back in one form or another, I think it was just a bit too early." — Mike Silagadze, CEO & Founder, ether.fi
Ethereum's restaking sector has lost approximately $13 billion in total value locked since its December 2024 peak. EigenCloud, the rebranded EigenLayer platform, holds $5.1 billion in TVL as of August 2026, down 77% from its $22.06 billion high recorded on August 14, 2025. Symbiotic, the nearest competitor, sits at $342.8 million, down 87% from its own $2.7 billion peak.
The largest liquid restaking protocol, ether.fi, has removed all restaking exposure from its flagship token weETH and plans to sever its last structural link to EigenLayer by Q4 2026. Puffer Finance's TVL fell from $1.3 billion to $62 million. The EIGEN token trades at $0.19, down 96.6% from its all-time high of $5.65 in December 2024. These are not isolated events. The entire restaking thesis — that passively restaked ETH could secure a marketplace of Actively Validated Services and generate sustainable yield — faces a structural revenue gap that points-based incentive programs temporarily concealed.
The restaking market peaked in late 2024 with sector-wide deposits exceeding $18.3 billion across liquid restaking protocols alone. EigenLayer's own TVL reached $22.06 billion in August 2025. The contraction since then has been severe and sustained.
| Protocol | Peak TVL | Current TVL (Aug 2026) | Decline | |----------|----------|----------------------|---------| | EigenCloud (EigenLayer) | $22.06B | $5.10B | -76.9% | | Symbiotic | $2.70B | $342.8M | -87.3% | | Karak | ~$1.0B | $102M | -89.8% | | ether.fi (restaking portion) | ~$6.2B | <$33M (<1% of $3.3B) | -99.5% | | Puffer Finance | $1.30B | $62M | -95.2% |
According to a Fensory market analysis from March 2026, EigenLayer's $8.9 billion TVL at that time represented approximately 10% of the total DeFi ecosystem's $94.08 billion. By August, even that share had contracted further. Liquid staking protocols such as Lido ($18.32 billion TVL) continued to outpace pure restaking solutions by a 2:1 margin, according to the same analysis.
The drawdown accelerated after points-based incentive campaigns ended. Capital that entered the restaking sector primarily for airdrop exposure rotated out as token generation events concluded without delivering sufficient returns to justify the added smart contract and slashing risk.
The most consequential development of 2026 for the restaking sector is ether.fi's systematic withdrawal from EigenLayer. The protocol, which manages approximately $3.3 billion in staking assets (down from a $12.43 billion all-time high in August 2025), was the single largest source of restaked capital on EigenLayer.
In early August 2026, ether.fi removed all restaking exposure from weETH, its flagship liquid staking token. Restaking functions were confined to a new, separate token called weETHs, built on Symbiotic — not EigenLayer. As of August 2026, less than 1% of ether.fi's assets remain restaked with EigenLayer, down from approximately half in early 2026. The protocol plans to completely remove EigenPod withdrawal credentials from its validators by Q4 2026, eliminating its last structural link to EigenLayer.
The decision reflects ether.fi's broader strategic pivot. Its "Summer" release, which went live across web, iOS, and Android on August 13, 2026, added trading in tokenized stocks and metals alongside crypto, an integrated Aave market on Optimism for portfolio-backed borrowing, and fiat on/off ramps covering more than 30 currencies. The company is repositioning as a crypto neobank rather than a restaking aggregator.
This pivot carries a clear economic signal: ether.fi's leadership concluded that restaking yield — in its current form — does not justify the complexity and risk relative to plain liquid staking or the revenue potential of neobank services.
The foundational promise of restaking was that Actively Validated Services would pay for shared Ethereum security, generating fee revenue that would flow to restakers as yield. The data shows this has not materialized at the scale the TVL implied.
According to a Fensory analysis, restaking protocols "rely heavily on token emissions" because AVS networks have been unable to "generate sustainable fee revenue." Most restaking yield in 2026 comes from EIGEN token emissions rather than from AVS-generated fees. The baseline EigenCloud reward rate stands at 3.87% annually, but the AVS-attributable portion of that yield remains unclear and likely small.
According to Tokenomics.com, EigenLayer commands a 93.9% market share of the restaking ecosystem with $15.258 billion TVL and 4,364,467 ETH (figures from an earlier measurement period), but revenue capture mechanisms remain nascent. The gap between capital committed and revenue generated defines the sector's central problem: restakers are being compensated primarily through inflation, not through fees from services that need security.
EigenDA, described by Coin Bureau as "the clearest live infrastructure product" in EigenLayer's portfolio, competes in the data availability market where Celestia commands approximately 50% market share after processing over 160 gigabytes of rollup data. Specific EigenDA revenue figures for 2026 are not publicly disclosed, limiting independent assessment of its commercial traction.
The liquid restaking token (LRT) landscape has consolidated rapidly. Ether.fi retains roughly 65% of LRT market share, but its shift away from restaking effectively converts that dominance from a restaking metric to a liquid staking metric.
Puffer Finance illustrates the severity of capital flight. The protocol launched in February 2024 with nearly $200 million in TVL on its first day and grew to over 500,000 ETH (worth more than $1.3 billion) by October 2024. Its TVL now stands at $62 million. The PUFFER token hit an all-time high near $1.00 in December 2024 and has since fallen to an all-time low of $0.025 — a 97.5% decline.
The EIGEN token itself trades at $0.19 with a market cap of approximately $137.9 million. Its all-time high was $5.65 in December 2024, placing the current price 96.6% below peak. The fully diluted valuation is approximately $386 million based on a total supply of 1.82 billion tokens, of which approximately 740 million are in circulation.
A scheduled token unlock on August 1, 2026, released approximately 36.82 million EIGEN tokens (about 2% of supply), valued at $7.63 million. These routine unlocks introduce predictable sell-side pressure in a market with limited buy-side demand.
In April 2026, Kelp DAO suffered a $292 million exploit — the largest DeFi hack of 2026 at the time — demonstrating the security risks inherent in cross-chain restaking infrastructure. According to CoinDesk reporting, the attacker exploited a 1-of-1 verifier configuration in Kelp's LayerZero cross-chain bridge adapter.
The attack was not a smart contract vulnerability in the conventional sense. Attackers compromised internal RPC nodes and launched a DDoS attack against external nodes, feeding false data to a single-point-of-failure verification network. This triggered the bridge to release 116,500 rsETH to an attacker-controlled address across 20 chains.
According to Halborn's post-mortem, security researchers attributed the attack to TraderTraitor, a North Korean state-sponsored threat group (also known as UNC4899), citing analysis from Mandiant, CrowdStrike, and other security firms.
The incident prompted structural fallout. According to CoinDesk, LayerZero acknowledged it "made a mistake," while Kelp shifted its rsETH bridge infrastructure to Chainlink. Solv Protocol moved more than $700 million in tokenized bitcoin infrastructure away from LayerZero in response.
The activation of on-chain slashing across EigenLayer marked a significant shift in the sector's risk profile. Previously, penalties for AVS misbehavior existed in principle but lacked full on-chain enforcement. With slashing now programmable and enforceable at the contract level, 33 slashing events were recorded across the EigenLayer ecosystem in Q1 2026.
The structural risk for restakers is compounded by double slashing exposure. When a validator restakes ETH into EigenLayer AVSs, the same capital becomes subject to slashing conditions on both the Ethereum consensus layer and the AVS layer simultaneously. A single misbehavior event, such as a double-sign, can trigger penalties on both layers.
EigenLayer implemented Unique Stake Allocation to isolate slashable stake across specific Operator Sets, limiting contagion. However, according to a Fensory analysis, cascading slashing events could still result in "significant principal losses beyond traditional staking risks." Slashing penalties are typically capped at a percentage of stake (not 100%), but the uncertainty around compounded scenarios remains a material consideration for institutional allocators.
EigenLayer's response to the revenue gap is twofold: tokenomics reform and product expansion beyond restaking.
ELIP-12, introduced in May 2026, proposes a 20% protocol fee on subsidized AVS rewards and directs 100% of EigenCloud infrastructure fees (after operator costs) into an EIGEN buyback contract. The proposal also establishes an Incentives Committee to direct emissions toward fee-generating AVSs, attempting to create deflationary pressure.
The broader strategic shift is the rebrand from EigenLayer to EigenCloud, positioning the platform as a "verifiable cloud" with three product lines: EigenDA (data availability), EigenCompute (verifiable off-chain computation, mainnet targeted for Q3 2026), and EigenVerify. An AgentKit beta launched in March 2026 as a developer toolkit for building crypto-native AI agents on EigenCloud.
The rebrand attracted a $70 million investment from a16z, according to Edgen Tech reporting. Whether cloud infrastructure revenue can fill the gap that AVS fees have not remains the open question for the protocol's long-term value proposition.
Ether.fi's decision to build weETHs on Symbiotic rather than EigenLayer signals a competitive dynamic. Symbiotic differentiates by accepting any ERC-20 token as collateral in isolated per-network vaults with network-defined slashing. Its core contracts are immutable and fully permissionless. Symbiotic has not yet launched a token.
According to data from Tokenomics.com and Protofire, the restaking market as of mid-2026 breaks down as: EigenLayer at 93.9% market share ($15.258 billion TVL), Symbiotic at 5.5% ($897 million), and Karak at 0.6% ($102 million). These figures vary across sources and measurement periods — more recent data from The Defiant places Symbiotic at $342.8 million and EigenCloud at $5.1 billion, reflecting continued contraction.
Babylon, focused on Bitcoin restaking, represents a separate vertical that does not directly compete for the same ETH capital but applies the shared-security thesis to a different base asset.
The restaking sector's trajectory from $18.3 billion in liquid restaking deposits to its current contracted state follows a pattern familiar across crypto infrastructure cycles: speculative capital enters on incentive programs, inflates TVL, and exits when the underlying revenue model cannot sustain returns. The core thesis — that shared security generates sustainable yield — remains unproven at scale.
EigenLayer's pivot to EigenCloud acknowledges this reality implicitly. The protocol is evolving from a restaking marketplace into a full-stack verifiable infrastructure provider, betting that cloud compute and data availability fees can generate the revenue that AVS demand alone has not. Whether ELIP-12's buyback mechanism, EigenCompute's mainnet launch, and the broader product expansion can reverse the capital outflow is the sector's defining question for the remainder of 2026.
The departure of ether.fi — the sector's largest client — from EigenLayer is not merely a competitive loss. It is a market signal that the largest allocators in the restaking ecosystem have concluded that, for now, the risk-adjusted returns do not justify the complexity. The restaking thesis has not been invalidated, but its timeline has been extended.