Restaking Without a Token: Symbiotic, Karak and OpenGDP

The Restaking Sector Has a Token Problem

On August 1, 2026, EigenLayer’s foundation is scheduled to unlock another 36.82 million EIGEN tokens to early contributors, the same size cliff that hit the market exactly a month earlier and worth roughly $7.6 million at current prices. It is a routine vesting event for a token that has already fallen more than 96% from its all time high, but it lands at an interesting moment for the restaking sector as a whole, per crypto.news. Two of EigenLayer’s most credible rivals have spent 2026 quietly answering a question EigenLayer itself never really got to ask before shipping its own token: does a restaking protocol actually need a liquid, tradeable token at all?

Symbiotic, the protocol most often described as EigenLayer’s closest technical competitor, has now run for more than two years without ever issuing one. Karak, once marketed as a universal, asset agnostic restaking layer, rebranded entirely in mid-2026 into something called OpenGDP that does not mention restaking anywhere on its own homepage. Meanwhile Babylon and SSV Network, the two protocols in this category that did launch tokens, have both watched their prices fall more than 90% from their respective peaks.

This piece looks at the non-EigenLayer half of the restaking map: why Symbiotic keeps delaying its token, what Karak actually became, and whether having a token has told investors anything useful about a restaking protocol’s health in 2026 at all.

What Restaking Actually Means

Restaking lets someone who has already staked ETH, or who holds a liquid staking token representing staked ETH, pledge that same capital a second time to help secure other services: oracles, cross chain bridges, data availability layers, rollup sequencers. In exchange for opting in, the staker earns extra yield on top of ordinary staking rewards; in return, the protocol can slash that stake if the extra service being secured misbehaves or gets compromised. EigenLayer coined the model and built the first mainstream implementation of it on Ethereum, and its earliest, most established AVS is still its own data availability layer, EigenDA, alongside independent services like Lagrange’s zero-knowledge state committees and the Omni Network interoperability hub. HOGE Wire has covered how liquid staking tokens became DeFi’s reserve collateral in the first place, which is worth understanding before restaking, since most restaking activity starts from an LST rather than raw staked ETH sitting in a validator.

The services being secured are usually called Actively Validated Services, or AVSs. A validator or an operator opts a unit of restaked ETH into one or more AVSs, and that unit of stake is now exposed to whatever slashing conditions that specific AVS defines, on top of ordinary Ethereum validator penalties. That stacking of obligations, one deposit securing several unrelated systems at once, is the core idea, and it is also where most of the sector’s risk concentrates. It is also why Vitalik Buterin warned back in 2023 that overloading Ethereum’s consensus with too many extra duties raises the costs, complexities and risks of running a validator, a caution the sector has spent the years since trying to design around rather than ignore.

EigenLayer Set the Template, Then Had to Defend It

EigenLayer, now operating under the EigenCloud brand, turned slashing on for mainnet operators on April 17, 2025, after roughly two years of deposits accumulating under an opt-in, unslashed regime. The rebrand pushes the platform toward verifiable cloud services, EigenDA for data availability, EigenCompute and EigenVerify for off-chain computation with on-chain proofs, rather than restaking security alone. A December 2025 governance proposal, ELIP-012, tried to align incentives with that shift by replacing the old emissions mechanism with a weekly EmissionsController that directs EIGEN rewards toward AVSs actively generating fees rather than idle deposits, while routing all cloud-service fees toward a buyback mechanism. Value-locked estimates for the EigenCloud stack are contested across analytics platforms and have ranged from the high single-digit billions to figures above $15 billion at different points in 2026 depending on what gets counted, a spread wide enough that no single figure should be treated as definitive, per market data tracked alongside CoinGecko’s EIGEN listing.

The token has not tracked that growth story regardless. EIGEN priced at an all time high of $5.65 on December 16, 2024, and traded at $0.1814 as of this writing, down 96.8% from that peak, with roughly $134.3 million in circulating market capitalization, per CoinGecko. The redesigned emissions model has not yet turned that around, and the August 1 unlock adds fresh supply into a token still searching for a floor. Whatever the cloud pivot accomplishes for EigenCloud’s product roadmap, it has not yet shown up in EIGEN’s price.

Symbiotic: $34.8 Million Raised, Still No Token

Symbiotic launched its mainnet with slashing live from day one in January 2025, a deliberate contrast with EigenLayer’s multi-year opt-in ramp. Its pitch is flexibility: almost any ERC-20 token can serve as restaked collateral, not just ETH or ETH-denominated liquid staking tokens, and each network built on Symbiotic can choose its own resolver for slashing disputes, a DAO committee, UMA, Kleros, rather than accepting one standardized arbitration model across the whole platform. The company has raised $34.8 million across two rounds: a $5.8 million seed co-led by Paradigm and Cyber Fund in June 2024, and a $29 million Series A led by Pantera Capital with Coinbase Ventures participating in April 2025, per CoinDesk.

What it has never done is turn its points program, running continuously since June 2024, into an actual token. Restakers have been earning Symbiotic Points for more than two years with, in the protocol’s own framing, zero contractual guarantee those points convert into anything at all; no confirmed supply, allocation, vesting schedule or conversion mechanics have ever been published, per FinanceFeeds. TVL estimates for Symbiotic vary about as widely as EigenCloud’s own figures: recent readings range from roughly $330 million on a narrow base-layer count up to figures above $1.6 billion once broader vault deployments and partner applications are included, a spread that says as much about how inconsistently this sector gets measured as it does about Symbiotic specifically.

On July 1, 2026, Symbiotic tried to change the framing entirely. Core V2 repositions the protocol away from restaking-for-AVS-security and toward what it calls shared collateral markets: idle vault capital automatically routes into lending venues like Aave and Morpho when it is not actively backing insurance, credit or tokenized-asset products, then recalls itself on-chain when obligations arise, with independent risk parameters enforced per vault. The company claims roughly 70% better capital efficiency than running standalone liquidity pools for each individual use case, and launched Liquid Lane, an instant-settlement product for tokenized real-world assets, with Midas as the first issuer and Fasanara Capital, a roughly $6 billion London-based asset manager, as an early curator, per The Block. Symbiotic co-founder Misha Putiatin framed the opportunity bluntly: “The RWA market has crossed $33 billion, but most of those assets still can’t be redeemed on demand. Institutions understand that, which is why liquidity gets priced at a premium,” he told FinanceFeeds. Nexus Mutual founder Hugh Karp, whose insurance protocol now leans on Symbiotic-backed capital as reinsurance capacity, described the structural logic this way: “Symbiotic enables delegated capital to sit behind Nexus Mutual as reinsurance capacity, with first-loss and second-loss exposure defined separately,” per The Block.

Why Symbiotic Keeps Delaying

Symbiotic has never publicly stated why it has not issued a token, but the pressure to do so is real and growing. FinanceFeeds frames the risk directly: every month that passes without a token announcement raises the odds that restakers simply rotate capital toward EigenLayer or Babylon, protocols that at least offer a liquid asset even if that asset has performed poorly since launch. Two and a half years after the points program started, with no supply schedule, allocation formula or conversion mechanics ever published, the silence has shifted from still early to a genuinely open question about whether a token is coming at all, or whether Core V2’s collateral-market pivot is Symbiotic quietly redefining itself as infrastructure that does not need one.

There is a more charitable reading too. EIGEN, BABY and SSV all launched tokens, and all three fell more than 90% from their respective highs regardless, so Symbiotic waiting until Core V2’s collateral-market products have real usage numbers to point to is not obviously the wrong call. A token launched into a falling market with no revenue story behind it just becomes one more chart trending toward zero, and Symbiotic has had a front row seat to watch that happen to three separate competitors. Waiting also sidesteps, for now, a regulatory question the rest of the sector has not resolved either, which is covered further down.

Karak’s Exit: From Universal Restaking to OpenGDP

Karak was built by Andalusia Labs, which raised a $48 million Series A in December 2023 led by Lightspeed Venture Partners and Mubadala Capital at a valuation above $1 billion, with the company simultaneously announcing plans for a global headquarters in Abu Dhabi Global Market as part of its international expansion, per FinancialIT. Karak marketed itself as a universal restaking layer: asset agnostic, supporting stablecoins, WBTC and ETH derivatives alongside plain ETH, and chain agnostic across roughly seven networks, with its own K2 chain intended as settlement infrastructure for the whole system. It never confirmed a KAR token listing on a major exchange; the only trading activity found for a token bearing that name has been thin, illiquid decentralized-exchange volume, not the kind of price a serious comparison should treat as a real market.

Sometime in mid-2026, Karak rebranded to OpenGDP. The old karak.network domain now redirects, and the new site describes itself as “the operating layer for real world economic execution,” unifying money, assets and autonomous systems for enterprises, institutions and governments, per a direct visit to opengdp.network. The homepage copy makes no reference to restaking, AVSs, or Karak’s old universal-collateral pitch anywhere in its main content. The footer still links out to legacy V1 Staking, V2 Staking and K2 Bridge pages, remnants of the old product that the company has not fully removed even as the front-facing brand has moved on completely, with community airdrop guides suggesting existing K2 users have been advised to unstake or bridge out.

What OpenGDP Actually Sells Now

The pivot is more than cosmetic rebranding. OpenGDP’s own language centers on three functions: moving value through stablecoin settlement, programming assets as tokenized instruments, and orchestrating economic workflows across institutions and autonomous systems, explicitly pitched at treasury and payments teams, institutional tokenized-asset desks, and government disbursement or procurement use cases. That is a materially different business than restaking ETH to help secure someone else’s oracle, and it puts Karak’s old restaking product in roughly the same bucket as infrastructure other protocols have quietly kept running while moving their actual narrative elsewhere: still technically operating, still tracked by some analytics platforms, but no longer the story the team is telling.

That gap between what trackers say and what the protocol says about itself is worth flagging on its own. Several third-party research pages still describe OpenGDP as a restaking or multi-asset-security product with an associated TVL figure in the tens of millions of dollars, a fraction of Symbiotic’s or EigenCloud’s totals and reportedly still falling. At least one aggregator has gone further and described OpenGDP as an AI compute and GPU marketplace business, a characterization that does not appear anywhere in OpenGDP’s own site copy as of this writing. The practical takeaway for anyone tracking this sector: a protocol’s positioning can move faster than the trackers that cover it, so a claim about what a project is now is only as reliable as how recently someone actually checked the primary source rather than a secondary aggregator’s cached description.

Three Different Answers to the Same Design Question

Strip away the branding and EigenLayer, Symbiotic and Karak’s original restaking product were all answering the same design question differently: who decides what counts as a slashable offense, and how flexible is the collateral allowed to be. EigenLayer’s answer, even post rebrand, is the most centered on Ethereum itself: restaked ETH and ETH LSTs, an Operator Sets model introduced in September 2024 that limits a given unit of stake to a single slashing authority at a time so one AVS’s failure cannot automatically cascade into another’s, and a foundation-run governance process for approving new AVSs.

Symbiotic’s answer is closer to a marketplace than a single system: any ERC-20 can be collateral if a given vault operator accepts it, and each network built on top chooses its own dispute resolver rather than inheriting one from the base protocol. That flexibility is also why Symbiotic’s TVL is harder to pin down than EigenCloud’s; it is less a single pool and more a collection of independently configured vaults. Karak’s original answer went a step further still, asset and chain agnostic by design, which was exactly the pitch that made it attractive early on and, arguably, exactly the scope that made it hardest to sustain a coherent security story around once EigenLayer’s Ethereum-native version proved out first. OpenGDP’s pivot away from that scope is one plausible resolution to a genuinely hard problem: rather than try to out-execute EigenLayer at securing arbitrary assets across arbitrary chains, build something adjacent instead.

Five Restaking Protocols Compared

Put together, the five protocols most commonly grouped under restaking in 2026 now occupy noticeably different positions, not just different market capitalizations.

Babylon and SSV Have Tokens. It Has Not Saved Them.

Babylon takes a different approach to restaking entirely. Instead of an EVM smart contract locking an LST, it uses Bitcoin’s own timelock scripting to lock native BTC directly, no wrapping or bridging required, and that locked BTC secures proof-of-stake networks that opt in as what Babylon calls Bitcoin Supercharged Networks. Its own dashboard has shown 56,853.16 BTC staked, worth roughly $5.64 billion, essentially unchanged for weeks across repeated checks, which looks less like a live counter and more like a periodically refreshed figure, worth flagging for anyone tempted to treat it as updating in real time, per Babylon’s own site. Bitcoin holders looking for on-chain yield without restaking at all have other options too, including supplying capacity directly to Lightning Network’s liquidity market, a comparison worth making before assuming Babylon-style BTC restaking is the only route to a return on otherwise idle Bitcoin.

SSV Network takes yet another angle: distributed validator technology, splitting a single validator’s signing key across multiple independent node operators so no single operator’s failure or downtime takes the validator offline. Its more recent Based Applications framework lets restaked and non-restaked capital alike secure additional services through that same distributed-operator infrastructure. For a deeper look at how DVT changes the risk profile of running a validator in the first place, HOGE Wire has covered client diversity, DVT and the four ETH staking path in detail.

Both protocols shipped a liquid, tradeable token early on, which restaking’s token skeptics might expect to correlate with stronger price performance than the tokenless alternatives get. It has not worked out that way. BABY is down roughly 93% from its April 2025 high, per CoinGecko; SSV is down more than 96% from its March 2024 high, per CoinGecko. A live token did not insulate either project from the same market-wide drawdown that hit EIGEN.

The EIGEN Test Case

If a token were simply good for a restaking protocol’s health, EigenLayer would be the strongest evidence for it: real, revenue-generating AVSs, a rebrand into cloud infrastructure, years of head start over every competitor on this list. Instead EIGEN sits at $0.1814, down 96.8% from its all time high, with another unlock due August 1 adding to circulating supply, per CoinGecko and crypto.news.

EigenLayer’s own founder, Sreeram Kannan, made a version of this argument before the token even launched. Speaking to CoinDesk in September 2023, he argued restaking was actually the more conservative bet, not the riskier one: “Anything that restaking can do, already liquid staking can do, so I view restaking as a lesser risk than liquid staking,” he said, framing restaking as “just one particular use case of liquid staking, but actually reducing the risk of that one particular use case,” per CoinDesk. Three years on, the argument reads less like reassurance about restaking’s safety and more like an early, honest acknowledgment that the category’s value proposition was always incremental, a bit more yield for a bit more risk on top of staking that was already happening, rather than a wholesale new asset class that obviously deserved its own multi-billion dollar token.

Three different token designs, three different launch dates, one shared outcome: none of them held value. That is the actual backdrop against which Symbiotic’s and OpenGDP’s no-token choices should be read, not as timidity, but as a reasonably rational response to what happened to the projects that went first.

Institutions Want the Yield, Not Necessarily the Token

There is a regulatory dimension to this too that gets less attention than the price charts. The SEC’s Division of Corporation Finance issued staff statements on May 29, 2025 covering protocol staking, and August 5, 2025 covering liquid staking, concluding that certain staking activities do not, by themselves, involve the offer and sale of securities. Both statements are explicit that they do not extend to restaking or liquid restaking variations, a named exclusion rather than an oversight, per the SEC’s own newsroom. Staff statements carry no binding legal force and can be withdrawn, but their silence on restaking specifically leaves the entire sector, tokenized or not, in a gray zone that plain staking has now partly exited.

That gray zone plausibly cuts both ways on the token question. A protocol offering a token that pays restaking-linked yield sits closer to the kind of structure regulators have shown willingness to scrutinize; a protocol offering collateral-market infrastructure without an associated yield-bearing token, Symbiotic’s Core V2 pitch, sits further from it. None of this is confirmed regulatory guidance either way, but it is a plausible reason institutional desks evaluating restaking exposure in 2026 have shown more interest in the underlying yield and collateral mechanics than in which project has the most liquid token. Exchanges built for institutional-grade execution and custody, the kind compared in HOGE Wire’s institutional trading test of Coinbase, Binance, Kraken and OKX, are increasingly the access point institutions actually use to get exposure to this sector, rather than holding governance tokens directly in a wallet they manage themselves.

What Would Actually Change the Calculus

A handful of concrete developments would tell us more about whether the token-versus-no-token split actually matters than another quarter of price charts:

• Whether Symbiotic converts its points program into an actual token, and how it prices that token relative to Core V2’s real collateral-market revenue rather than pure emissions.
• Whether OpenGDP’s enterprise and government pitch produces a named customer or disbursement pilot that can be independently verified, rather than remaining a homepage description.
• Whether the Babylon Labs and Aave Labs partnership for native BTC collateral on Aave V4, announced in December 2025 and still not live as of this writing, actually reaches mainnet, since it would be the first serious bridge between Bitcoin restaking and mainstream DeFi lending.
• Whether EigenLayer’s draft proposal for a clean, irreversible restaking exit changes how capital behaves around future unlocks, a mechanic HOGE Wire has covered in a dedicated piece on restaking’s exit problem.

None of those four is guaranteed to resolve soon, and the sector has shown a consistent pattern in 2026 of announced integrations sliding past their original target dates. For now, the most honest summary is that restaking’s token question remains genuinely unresolved, not because nobody has tried an answer, but because every answer tried so far, EIGEN’s token, BABY’s token, SSV’s token, Symbiotic’s silence, Karak’s exit, has produced a project trading well below where it started.

Frequently Asked Questions

What is restaking in crypto?

Restaking is the practice of using ETH that is already staked, or a liquid staking token that represents staked ETH, to secure additional blockchain services on top of ordinary Ethereum validation. Those additional services, often called Actively Validated Services or AVSs, can include oracles, cross-chain bridges, data availability layers and rollup sequencers. In exchange for opting in, the staker earns extra yield, but also accepts new slashing conditions specific to whatever service they are helping secure. EigenLayer introduced the model on Ethereum; Babylon applies a similar idea to native Bitcoin.

Does Symbiotic have a token?

No. As of July 2026, more than two years after Symbiotic began running a points program in June 2024, the protocol has not launched a public token, disclosed a supply schedule, or confirmed a launch date. Restakers earn Symbiotic Points with no contractual guarantee those points convert into a future airdrop or token. Symbiotic has instead focused recent development on Core V2, a shared collateral markets product launched July 1, 2026, rather than on a token launch.

Is Karak still a restaking protocol?

Not in how it presents itself. Karak, built by Andalusia Labs, rebranded to OpenGDP in mid-2026 and now describes itself as an operating layer for real world economic execution focused on stablecoin settlement and tokenized assets for enterprises, institutions and governments. Its homepage no longer mentions restaking, and legacy Karak staking products only remain accessible through footer links rather than being part of the current pitch, even though some third-party trackers still classify it under restaking with a declining TVL figure.

Is restaking safe?

Restaking carries more risk than plain staking, not less, because the same unit of stake can be exposed to slashing conditions from every service it secures at once. Academic research on correlated slashing has modeled how a bug or bad operator in one service can cascade losses across a staker’s entire restaked position. Protocols have responded with safeguards, including EigenLayer’s Operator Sets model, which limits a given unit of stake to a single slashing authority at a time, and third-party insurance products from providers like Nexus Mutual. None of that makes restaking risk-free; it remains a higher-yield, higher-risk layer on top of ordinary staking.

What is the difference between staking and restaking?

Staking means locking a cryptocurrency, most commonly ETH, to help validate its own base blockchain in exchange for that network’s native reward rate. Restaking means taking capital that is already staked, or a liquid token representing it, and pledging it a second time to secure separate, unrelated services for additional yield. Staking has one set of slashing rules, defined by the base network; restaking adds a second, protocol-specific layer of slashing risk on top, one for each additional service the staker opts into.

Elena Marsh, HOGE Wire