Babylon

Babylon lets bitcoin holders lock BTC in a script on Bitcoin itself, keep their keys, and use that locked coin to secure proof-of-stake chains in exchange for BABY rewards. Babylon bitcoin staking has run since August 2024, and about 40,800 BTC was staked on 14 October 2026. That same day, Babylon Labs asked the Aave lending protocol to accept native BTC as collateral on Aave V4.

The request is an ARFC (Aave Request for Final Comment) posted on the Aave governance forum at 14:26 UTC. It is a proposal and nothing more yet. No vote has been held on it, risk parameters are not set, and the vault system it depends on runs only on a public testnet.

How Babylon bitcoin staking works

Bitcoin has no staking of its own. It runs on proof of work, and nothing in the protocol pays holders for locking coins. Babylon uses the word for something else: BTC is locked as collateral that can be partly burned if the validator it backs cheats on another chain. That threat of loss gives the BTC its security value, and the chain being secured pays for it in its own token.

The first chain secured this way is Babylon Genesis, Babylon’s own network: a Cosmos SDK chain with CometBFT consensus, CosmWasm contracts and blocks of about 10 seconds, according to the project documentation. Its mainnet and the BABY token went live on 10 April 2025.

The coins never leave Bitcoin. They are not wrapped into an ERC-20 and not handed to an exchange. The holder keeps self-custody within the rules of the script they signed, which is where Babylon differs from most “bitcoin yield” products: those start by turning BTC into a token on another chain.

The script and the slashing rule

A Babylon stake is a Bitcoin transaction that sends BTC to a script with three spending paths, as Babylon’s documentation describes it. The holder can spend alone once the timelock expires. The holder can unbond early, which returns the coins after a shorter delay. The third path is slashing: it needs signatures from a covenant committee and lets part of the stake be burned.

Stakers do not run validators. They delegate to a finality provider, an operator that votes on Babylon Genesis blocks with extractable one-time signatures (EOTS). The provider commits in advance to public randomness for each future block height. If it signs two different blocks at the same height, it reuses that randomness, and its private key can be computed from the two signatures. Its voting power drops to zero and the exposed key can sign the slashing transactions of every delegation behind it.

So the staker’s money is at risk in one narrow case: the provider they picked double-signs. On 14 October 2026 the staking API counted 20 active finality providers out of 133 registered, and 10,392 active delegations.

Staking parameters in October 2026

Values from the Babylon Genesis chain (parameter version 5), read on 14 October 2026:

Parameter	Value
Stake per delegation	0.005 BTC to 5,000 BTC
Staking timelock	64,000 Bitcoin blocks (about 444 days, our computation at 10 minutes a block)
Unbonding time	301 blocks (about 50 hours); versions 0 to 4 used 1,008 blocks, about 7 days
Slashing rate	0.1% of the stake
Covenant committee	9 keys, 6 needed to sign
Minimum finality provider commission	3%

A 0.1% slashing rate costs 0.001 BTC on a 1 BTC stake. Time costs more: the stake is locked for 64,000 blocks unless the holder unbonds, and locked BTC cannot be sold or moved. One Babylon documentation file still gives an unbonding period of about 21 days, but the chain applies 301 blocks, in line with the “about two days” Babylon itself quoted in July 2025.

Kraken and Ledger: staking without a Bitcoin script

Many holders reach Babylon through an exchange. Kraken started BTC staking via Babylon in June 2025, paying rewards weekly in BABY and quoting “up to” 1% a year at launch, for clients in the US (minus seven states, New York and California among them), the UK, Australia and the UAE. The Block reported a seven-day unbonding period on Kraken, with no rewards during it. Those terms date from 2025; check Kraken’s current ones in your country.

Ledger added Clear Signing for Babylon’s bitcoin vaults in March 2026. Support for BTC staking itself was announced as “planned”.

Babylon BTC staking rewards: paid in BABY, not bitcoin

Stakers earn BABY, not BTC. The rewards come from new BABY issuance, set at 5.5% a year since the v4 upgrade passed on 13 November 2025 (it was 8%). They go to BTC stakers, BABY stakers and “co-stakers” who lock both, with a small share for finality providers and validators. Most proof-of-stake chains pay staking rewards the same way. The difference here: the asset at risk (BTC) is not the asset paid out.

Babylon publishes no fixed APR. At 18:30 UTC on 14 October 2026 its staking API showed a field called “max_staking_apr” at 0.545% and another, “btc_staking_apr”, at 0.0369%. Neither is documented. Whatever the rate, its dollar value depends on the BABY price on the day the rewards are sold.

Babylon and Aave V4: the native BTC collateral proposal

The ARFC, titled “Onboard Native BTC as Collateral to Aave V4 via Babylon Trustless Bitcoin Vaults”, asks Aave governance to deploy two new Spokes. In Aave V4, live on Ethereum mainnet since 30 March 2026, a Spoke is a market that draws liquidity from a shared Hub. The first use would be borrowing stablecoins against native BTC.

• Babylon Spoke: its only collateral is native BTC locked through Babylon’s Trustless Bitcoin Vaults (TBV). It is a variant of the standard Spoke, written by Aave Labs in the Aave V4 repository.
• BTC Vault Swap Spoke: custom code that settles positions after a liquidation.

Babylon Labs puts its case in one line of the proposal: “The deposited BTC is not wrapped, bridged or held by a custodian.” Inside Aave, each vault is counted by an accounting token with 8 decimals that cannot be transferred, traded or held outside the integration. The token contract has no admin and cannot be upgraded.

Deposits and redemptions

The depositor locks BTC in a Taproot output on Bitcoin. Getting it back takes a zero-knowledge proof of the right event on the host chain, Ethereum first, and a false claim can be challenged during a fraud-proof window. That window is why the ARFC says redemption takes “multiple days”. A borrower who repays on Ethereum on a Monday should not expect the coins back on Bitcoin that day.

On top of Aave interest, the adapter contract can charge a Babylon interest fee, a percentage of the borrower’s base Aave interest.

Liquidations are settled in WBTC

Liquidation takes one whole vault at a time. Anyone can liquidate. A liquidator who cannot redeem on Bitcoin swaps the seized vault through the Swap Spoke for WBTC, at the oracle rate minus a discount. A permissioned set of “arbitrageurs” later buys the escrowed vaults and redeems the BTC on Bitcoin. If a vault no longer covers its debt, the shortfall is booked to the Hub as a deficit.

Native BTC goes in, then, but a liquidation is settled in a wrapped token held by a custodian. If a borrower’s debt is cleared before the whole vault is used up, the ARFC says the borrower gets the rest as discounted WBTC, not native BTC.

Who controls what

• Aave DAO governs both Spokes: caps, rates, oracle, freeze and pause.
• A 2-of-3 multisig of Llama Risk, Babylon Labs and Aave Labs holds a separate configuration role on the Swap Spoke.
• “Babylon governance representatives” run the TBV contracts through a public timelock. Their freeze blocks new vaults and new borrows but leaves exits open; their pause also stops exits and is kept for emergencies.

Revenue sharing between Aave and Babylon

The split covers “Qualifying Revenue” only: the Hub’s liquidity fee on Babylon Spoke interest and the liquidation fees, not all interest paid by borrowers. The first $500,000 each year goes entirely to Aave DAO.

Yearly qualifying revenue	Aave DAO	Babylon	Token purchases
First $0.5M	100%	0%	0%
$0.5M to $2M	78%	18%	4%
$2M to $4M	72.5%	22.5%	5%
$4M to $6M	67%	27%	6%
Above $6M	61.5%	31.5%	7%

The token-purchase slice buys AAVE (85%) and BABY (15%) on the open market. For BABY holders it is the only written link between the vault business and the token so far. Babylon’s October 2025 roadmap said network fees “could be used” to buy and burn BABY: an option, not a commitment.

Where it stands: proposal and testnet

One vote has passed. A TEMP CHECK went up on the Aave forum on 25 May 2026, Aave Labs backed it on 1 June (“BTC collateral demand is real”, it wrote), and the Snapshot vote of 4 to 7 June closed with 299,743 votes for and none against. The ARFC is the next stage.

The proposal lists what remains: community feedback, a Snapshot vote on the ARFC, then an AIP with full risk parameters for on-chain approval. Aave’s risk providers have not published loan-to-value ratios or caps, and no mainnet date exists. The Defiant also reported that risk settings and further votes are pending. On Babylon’s side, the borrowing flow through Aave V4 runs on its public testnet with test assets, as described on 25 September 2026.

The calendar has slipped once. When the two companies announced their partnership on 3 December 2025, they targeted a joint launch in April 2026, pending Aave governance. Stani Kulechov, founder of Aave Labs, said then that the plan “brings native BTC into DeFi without wrappers or centralized custody.”

Trustless Bitcoin Vaults: the cryptography behind the Aave plan

Bitcoin cannot run smart contracts, so a loan on Ethereum cannot seize coins on Bitcoin directly. TBV tries to close that gap with cryptography. Babylon published the whitepaper in August 2025 and, in October 2025, borrowed USDC against native BTC on Morpho on Ethereum mainnet as a first live test.

The engine is BaBe, from the paper “BABE: Verifying Proofs on Bitcoin Made 1000x Cheaper” (IACR ePrint 2026/065, received 16 January 2026, last revised 5 October 2026). Its authors are Sanjam Garg and Dimitris Kolonelos (UC Berkeley), Mikhail Sergeevitch (Babylon Labs), Srivatsan Sridhar and David Tse. It builds on BitVM3 and garbled circuits, and the ePrint page lists ACM CCS 2026 as its venue.

According to the ARFC, six firms audited parts of TBV, with overlapping scopes:

Firm	Scope
Coinspect	Pre-audit, btc-vault, provers, contracts, web services
Sherlock	btc-vault, provers, contracts, web services, BaBe
Zellic	btc-vault, contracts
ABDK	Provers, contract cryptography, BaBe
ZK Security	BaBe
Runtime Verification	Formal verification of the contracts

Coinspect, ABDK and Zellic are re-auditing the final code “before mainnet deployment”, and Babylon says those reports will be shared once finalized. The ARFC does not link the TBV reports. Each vault also depends on actors that must pre-sign it: the arbitrageurs are a permissioned set, the “universal challengers” a global one, and the ARFC names neither. A depositor can run their own vault provider and challenge bad claims themselves.

Not everyone accepts the “trustless” label. In an October 2025 Substack post, analyst Yash Belavadi argues that the vaults rely on a garbled verifier built off-chain and checked by cut-and-choose, which gives statistical confidence, not proof, and that Bitcoin cannot verify that setup. It is one analyst’s reading, but it points at where the trust sits.

Other partners have lined up. GoMining announced on 5 May 2026 a plan to activate up to 1,000 BTC through TBV. HashKey Cloud said on 7 October 2026 it would integrate TBV, with timing details to come “in the coming months”. Co-founder Fisher Yu said the design lets regulated institutions “use native Bitcoin as collateral and retain title.”

Staked BTC is down a third from its December 2025 peak

Babylon is the largest BTC staking protocol tracked by DefiLlama, which files it under “Restaking”. At 17:29 UTC on 14 October 2026, DefiLlama counted 40,772.44 BTC staked, worth about $3.34 billion at that time. The Babylon staking API gave the same BTC figure. The peak was 61,771 BTC on 14 December 2025, so the stake has fallen about 34% since; in dollars the high was $7.14 billion on 7 October 2025.

Date (DefiLlama)	BTC staked
1 Nov 2024	23,889
1 Jan 2025	56,979
10 Apr 2025 (Genesis launch)	49,797
1 Oct 2025	58,272
14 Dec 2025 (peak)	61,771
1 Jan 2026	60,994
1 Mar 2026	49,626
1 Jul 2026	51,385
1 Aug 2026	41,355
1 Sep 2026	39,099
1 Oct 2026	40,481
14 Oct 2026, 17:29 UTC	40,772

The sharpest drop came in late July 2026, when about 9,000 BTC left between 25 and 31 July (51,352 to 42,490 BTC). We found no source that explains it.

Babylon’s own figures are higher. In the ARFC, Babylon Labs says it has activated “over 100,000 BTC cumulatively” and peaked at 72,000 BTC. DefiLlama’s series shows no such peak, and Babylon does not explain the gap.

BABY token: supply, inflation and unlocks to 2029

BABY launched on 10 April 2025 with 10 billion tokens and has no hard cap. On 14 October 2026 the on-chain supply was about 11.02 billion BABY, and CoinGecko listed about 4.9 billion as circulating. BABY holders vote on chain governance; BTC stakers do not.

The allocation comes from the official tokenomics page, which notes that category shares may change, though never in favour of the team, investors or advisors.

Allocation	Share	BABY	Release
Early private investors	30.5%	3.05B	36 monthly unlocks, May 2026 to April 2029
Ecosystem building	18%	1.8B	25% at launch, the rest linear over 3 years from the first anniversary
R&D and operations	18%	1.8B	25% at launch, the rest linear over 3 years from the first anniversary
Community incentives	15%	1.5B	Fully unlocked
Team	15%	1.5B	36 monthly unlocks, May 2026 to April 2029
Advisors	3.5%	0.35B	36 monthly unlocks, May 2026 to April 2029

The first monthly unlock for investors, team and advisors fell on 10 May 2026. Those three groups hold 4.9 billion BABY. Over 36 months that is about 136 million BABY a month, roughly 1.36% of the initial supply, until April 2029. This is our computation from the official schedule; trackers publish different monthly figures.

Add 5.5% yearly issuance on top. Rewards paid in a token with that much new supply need buyers. This page makes no Babylon price prediction: the supply schedule is the part that can be checked.

At launch, 600 million BABY (6% of the initial supply) went to early participants in an airdrop, and Binance opened spot trading on 10 April 2025.

From a Stanford research project to Aave

Babylon was started in 2022 by David Tse, a Stanford professor, and Fisher Yu. Bitcoin staking opened in three capped rounds in 2024, before Babylon Genesis existed. Babylon’s “2024 in Review” gives the figures:

Date	Phase 1 round
22 Aug 2024	Cap 1: 1,000 BTC from 12,720 stakers
8 Oct 2024	Cap 2: 22,891 BTC from 12,590 stakers
10 to 17 Dec 2024	Cap 3: 33,399 BTC from 109,980 stakers; Phase 1 total 57,290.61 BTC

BSNs, or “Bitcoin Supercharged Networks”, were meant to come next: outside chains renting Babylon’s BTC security. The first batch, announced on 16 July 2025, named Osmosis, TAC, Union, Sui, BirdLayer, Corn, NodeOps, Plume, BOB and Manta Network. Then the “Vault First” roadmap of 7 October 2025 moved BSNs and an EVM on Babylon Genesis to after the vault launch, which was aimed at “early next year”. That launch has not happened.

The chain still ships upgrades. Governance proposal 24, described as a v4.5 upgrade with “important security fixes”, passed with voting ending on 14 September 2026, and release v4.5.0 was published on 29 September.

Team, funding and audits

The two public figures are the co-founders, David Tse and Fisher Yu. We found no official page listing other executives.

The disclosed rounds add up to about $111 million, close to the “approximately $110M+” Babylon cites in the ARFC:

• an $8 million seed round, disclosed in March according to The Block, which gave no day;
• $18 million Series A on 7 December 2023, co-led by Polychain Capital and Hack VC, with Framework Ventures, Polygon Ventures, Castle Island Ventures, OKX Ventures and Symbolic Capital among others;
• $70 million on 30 May 2024, led by Paradigm, with Bullish Capital, Polychain, HashKey Capital, Mantle, Galaxy, Hack VC, ViaBTC Capital, Amber and HTX Ventures among others;
• $15 million from a16z crypto on 7 January 2026, for Trustless Bitcoin Vaults.

The Block also reported an investment of undisclosed size from Binance Labs.

Babylon’s audit page links reports on the staking system and the chain from Zellic, Coinspect, Sherlock, Oak Security, Informal Systems and Halborn, dated from June 2024 to January 2026. We found no report of lost staked BTC since the 2024 launch.

The chain has had outages. Babylon Genesis halted on 18 June 2025; the GitHub advisory linked to that post-mortem describes a message that changed the validator set at an epoch boundary and made the chain panic, fixed in version 2.1.0. A second advisory, published on 28 June 2025, covered fees in a non-native token that could crash the chain (fixed in 2.2.0).

Babylon compared with EigenLayer and BTC liquid staking tokens

Babylon is often set beside Ethereum restaking: both reuse a staked asset to secure other networks. On EigenLayer, though, the ETH already sits on the chain that runs the contracts. Babylon has to prove events between Bitcoin and other chains, which is why it leans on timelocks, covenant signatures and, for vaults, fraud proofs.

Several liquid staking tokens are built on top. Babylon’s “2024 in Review” counted 12 such protocols, Lombard and Lorenzo among them. Lombard deposits BTC into Babylon staking and a consortium of institutions mints LBTC on Ethereum; Solv offers a Babylon version of SolvBTC. These tokens trade and work in DeFi, but they put a custodian or a consortium between the holder and the BTC, a layer native staking does not have. On 14 October 2026 DefiLlama listed Lombard LBTC at $641 million and SolvBTC at $521 million, against $3.34 billion for Babylon.

Risks of Babylon bitcoin staking

• Slashing: 0.1% of the stake if your finality provider double-signs. You carry its mistakes, so the choice of operator matters.
• Lock time: 64,000 blocks, with about 50 hours to unbond under current parameters. Locked BTC cannot be sold.
• The covenant committee: 6 of its 9 keys sign on the slashing path, a trust assumption outside Bitcoin itself.
• Reward value: BABY is inflationary and faces monthly unlocks until April 2029.
• Chain risk: Babylon Genesis halted in June 2025, and bugs in the chain or the staking software can delay exits.
• Vault risk: TBV and BaBe are new cryptography, not on mainnet, with permissioned arbitrageurs and liquidations settled in WBTC.

What to watch next

Native BTC lending on Aave needs four things that have no date yet: a Snapshot vote on the ARFC, an AIP with published loan-to-value ratios and caps, the TBV re-audit reports from Coinspect, ABDK and Zellic, and a mainnet launch. Meanwhile, about 136 million investor, team and advisor BABY reaches the market every month until April 2029.