Bitcoin Layer 2s are having a moment. Citrea just launched the first production ZK rollup on Bitcoin. BitVM is enabling fraud proofs without consensus changes. Stacks is maturing into real DeFi infrastructure. Lightning hit an all-time high capacity of 5,637 BTC.
But here’s the uncomfortable truth: total Bitcoin L2 TVL has dropped 74% from its peak. Only 0.46% of all Bitcoin sits in Layer 2s. Node counts are declining even as institutional capital flows in.
So what’s actually happening? Let’s break down the landscape.
Lightning Network: Payments Solved, But Limited
Lightning is the OG Bitcoin L2, and it works. The numbers are impressive:
- Capacity: 5,637 BTC (~$490M) - all-time high
- Success rate: 99.7% in controlled deployments
- Settlement: Under 0.5 seconds
- Cost savings: 80%+ vs on-chain payments
But here’s what the Lightning bulls don’t tell you: node counts dropped from 20,700 in 2022 to ~14,940 today. The network is becoming more capitalized but less decentralized. Institutional players are driving growth, not grassroots adoption.
Lightning solves payments. It doesn’t solve programmability. You can’t build a DEX, a lending protocol, or a DAO on Lightning. For that, you need something else.
Stacks: The Smart Contract Pioneer
Stacks has been building smart contracts on Bitcoin since before it was cool. Today it’s the largest application layer with:
- TVL: ~$200M+
- Developers: 150 monthly active
- Apps: DEXs, lending, staking, NFTs
The Nakamoto upgrade improved Bitcoin finality, making Stacks transactions settle faster. But Stacks has a fundamental challenge: it’s a sidechain with its own consensus and token (STX). Purists don’t consider it a “real” Bitcoin L2 because it introduces additional trust assumptions.
BitVM: The Breakthrough Nobody Saw Coming
BitVM changed the game. It enables fraud proofs on Bitcoin without any consensus changes—the holy grail for L2 development.
Here’s how it works: computations happen off-chain, but anyone can challenge false claims on-chain. It’s like optimistic rollups on Ethereum, but using Bitcoin’s existing script capabilities.
Projects building on BitVM:
Bitlayer: $788M TVL, launched April 2024. The current leader in BitVM-based scaling.
BOB: $131M TVL, currently an OP Stack rollup on Ethereum but evolving toward BitVM for Bitcoin security.
The limitation? BitVM is complex. The fraud proof mechanism requires sophisticated cryptographic commitments. It’s a framework, not a turnkey solution.
Citrea: The ZK Rollup Bet
On January 27, 2026, Citrea launched the first production-grade ZK rollup on Bitcoin. This is significant.
What Citrea offers:
- Full EVM compatibility (zkEVM)
- ctUSD stablecoin backed by US Treasuries
- BTC-collateralized lending without custodians
- BitVM-based bridging for trust minimization
Early metrics show promising activity—testnet reached 10% of Bitcoin’s monthly bandwidth at peak. The launch includes 40+ dApps and backing from Founders Fund, Galaxy, and Maven11.
But ZK proofs are computationally expensive. The question is whether the benefits justify the overhead compared to simpler approaches.
The Trust Assumption Spectrum
Not all L2s are created equal. Here’s how they compare on trust:
| Approach | Trust Assumption | Trade-off |
|---|---|---|
| Lightning | Watchtowers, channel partners | Limited to payments |
| Stacks | Separate consensus, STX staking | Full smart contracts, not Bitcoin-secured |
| BitVM (Bitlayer) | 1-of-N honest verifier | Complex fraud proofs |
| ZK Rollup (Citrea) | Math (ZK proofs) | Computational overhead |
| Sidechains (Liquid) | Federation | Fast but federated |
The purest approach (Lightning) has the most limited functionality. The most functional approaches (Stacks, Citrea) require additional trust assumptions.
The TVL Paradox
Here’s what puzzles me: VanEck predicts a $24B market cap for Bitcoin L2s. Institutional money is flowing in. Yet TVL dropped 74% from peak.
What’s happening?
- Speculation vs. utility: Much of 2024’s TVL was farming points and airdrops. Real usage is lower.
- Fragmentation: 50+ Bitcoin L2 projects competing for the same liquidity
- Cultural resistance: Many Bitcoiners actively oppose L2 complexity
- Bridging friction: Moving BTC to L2s is still clunky
What “Scaling” Actually Means
Different L2s are solving different problems:
- Lightning scales payments. Success metric: transaction throughput, cost per payment.
- Stacks/Citrea scales programmability. Success metric: DeFi TVL, developer activity.
- BitVM scales trust minimization. Success metric: security guarantees, decentralization.
There’s no single winner because there’s no single problem. Bitcoin might end up with a multi-layer ecosystem like Ethereum—different L2s for different use cases.
My Take
The Bitcoin L2 landscape is at an inflection point:
- Lightning will dominate payments. The 30% of BTC transfers projection by 2026 seems realistic.
- BitVM-based solutions will win for trust-minimized programmability. Bitlayer’s $788M TVL shows demand.
- ZK rollups (Citrea) are the long-term bet if compute costs drop.
- Stacks will remain the ecosystem leader until BitVM matures.
But the real question is whether Bitcoin needs to scale programmability, or if the base layer’s store-of-value function is enough.
Discussion questions:
- Which Bitcoin L2 approach will dominate by 2028?
- Is the TVL decline a sign of maturation or failure?
- Should Bitcoin even try to compete with Ethereum for DeFi?
- What’s the killer app for Bitcoin L2s beyond payments?
layer2_larry
Fast, cheap payments (Lightning)
DEXs, lending, DAOs (let Ethereum handle this)