Lombard (LBTC)

LBTC is, per lombard-finance/evm-smart-contracts, "liquid Bitcoin; it's yield-bearing, cross-chain, and 1:1 backed by BTC". It is a consortium-issued wrapped BTC, not a Bitcoin-enforced construct.

docs.lombard.finance advertised "$3 billion in Bitcoin onboarded" (cumulative, not outstanding) and integration with "Aave, Morpho, and 50+ DeFi protocols across 10 blockchains" when fetched on 15 September 2026. DefiLlama put Lombard's TVL at ~$906.7M on the same date — re-check before quoting: https://defillama.com/protocol/lombard

The yield source changed — read this first

LBTC launched as Babylon-staked BTC and that is how most older write-ups still describe it. It is no longer true.

The docs state a 2.5% net APY target in BTC terms, explicitly variable and "not a guarantee". Lombard's docs give only "2026" for the Bitwise switchover, not a day — do not invent one. Source: https://docs.lombard.finance/use/lbtc/understanding-yield

The security implication is the point: LBTC yield is now option premium sold by a regulated asset manager, with a short-volatility risk profile, not PoS staking rewards. A holder is exposed to capped upside in a sharp BTC rally.

Architecture

• Lombard Ledger — a Cosmos-based appchain running CometBFT, operated by the Security Consortium. Every deposit, mint, redemption and cross-chain transfer is recorded here before any token appears on a destination chain.
• Security Consortium — 14 institutional members; every action needs signatures from 10 of 14 (two-thirds supermajority). Per the members page (last updated 26 August 2026): Galaxy, OKX, Kraken, DCG (crypto institutions); Amber, Wintermute (market makers); Antpool, F2Pool (mining pools); Bitwise Onchain Solutions, Figment, Kiln, P2P (validators); Cubist, Nansen (security/research).
• CubeSigner — keys generated and held inside HSMs; they never leave secure hardware. No consortium member, Lombard employee or Cubist engineer can extract them.
• Bascule drawbridge — an independent verification layer run by Cubist. On deposit it confirms the BTC exists before mint; the Reverse Bascule confirms LBTC was actually burned before a BTC payout is authorised.
• Bridging — Chainlink CCIP, requiring authorisation from both the Consortium and Chainlink.

Mint / redeem

Mint:   BTC to consortium address -> relayer observes -> consortium
        verifies (6 confirmations) -> Bascule confirms -> contract
        mints LBTC
Redeem: burn LBTC -> Ledger records -> Bascule verifies the burn ->
        CubeSigner authorises the BTC payout

Contract surface

From the evm-smart-contracts README:

Non-EVM deployments exist too: sol-svm-contracts, sui-move-contracts.

Use cases

• BTC collateral in EVM DeFi with a yield accrual built into the exchange rate rather than a rebase.
• Cross-chain BTC without fragmenting liquidity per chain.
• Custody-adjacent products — Lombard's docs claim LBTC powers Bitcoin products for Ledger, Binance and Bybit users.

Limitations

• This is a federation. 10-of-14 named institutions can mint. The threshold is high and the members are reputable, but the trust model is closer to wBTC-with-more-signers than to tBTC's random group selection or a BitVM bridge.
• HSM custody is still custody — CubeSigner prevents key extraction; it does not prevent a quorum from signing.
• Yield is directional risk, not protocol revenue. See above.
• Bridge dependency — LBTC on a non-Ethereum chain additionally depends on CCIP or LayerZero.
• Do not describe LBTC as "Babylon liquid staking" in anything written after 2025 without re-checking the yield page.

See also

• babylon/SKILL.md
• solv/SKILL.md
• threshold-tbtc/SKILL.md
• core-dao/SKILL.md