200 Million Smart Accounts Deployed, 87% of UserOps Are Paymaster-Sponsored, and EIP-7702 Lets Any EOA Go Gasless - Account Abstraction Just Hit the Tipping Point

General
1

Account abstraction has been “the future of crypto UX” for three years. I want to make the case that in early 2026, it has stopped being the future and started being the present. The numbers are unambiguous, and the infrastructure stack is finally production-grade.

The Growth Numbers Are Staggering

Smart account deployments crossed 200 million in 2024-2025, representing a 5x year-over-year increase. To break that down:

  • Safe deployed 28.7 million accounts in 2024 alone, a 6.5x increase over 2023
  • 19.7 million ERC-4337 native smart accounts were deployed, a 7x increase over 2023
  • 103 million User Operations (UserOps) were executed in 2024, more than 10x the 8.3 million in 2023
  • 87% of all UserOps were paid by Paymasters, meaning the vast majority of smart account users never touched ETH for gas

That last statistic is the most important one. When 87% of transactions are gas-sponsored, the “users need to buy ETH to use crypto” problem is effectively solved for applications that implement paymasters. The UX has shifted from “figure out gas” to “just use the app.”

The Infrastructure Stack Is Now Production-Ready

Three years ago, building gasless transactions required stitching together experimental components from multiple providers. Today, the stack is mature:

ERC-4337: The Standard

ERC-4337 went live on Ethereum mainnet in March 2023 and has since expanded to every major L2. The architecture separates transaction validation from execution through UserOperations, Bundlers, and an EntryPoint contract. The key innovation is that Paymasters – smart contracts that sponsor gas – are a first-class concept in the protocol, not a hacky workaround.

EIP-7702: The Game Changer

The Pectra upgrade (May 2025) introduced EIP-7702, which allows any Externally Owned Account (EOA) to temporarily delegate execution to a smart contract. This eliminates the biggest adoption barrier for account abstraction: users no longer need to migrate from their existing MetaMask wallet to a new smart wallet. Their existing address can behave like a smart account for a single transaction and revert to normal afterward.

Circle has already built on EIP-7702 to enable gasless USDC transactions directly from standard EOAs. No smart wallet deployment required, no address change, no asset migration. This is the missing piece that makes gasless UX accessible to the 100+ million existing MetaMask users.

The Bundler and Paymaster Landscape

The infrastructure providers have matured into a competitive market:

  • Coinbase handled over 30 million UserOps in 2024, making it the largest bundler by volume
  • Pimlico provides bundler, verifying paymaster, and ERC-20 paymaster infrastructure
  • Biconomy offers cross-chain account abstraction with Wormhole and Axelar integrations
  • Alchemy operates bundler and paymaster infrastructure across multiple chains
  • Gelato provides paymaster and bundler compatible with both ERC-4337 and EIP-7702, now live on 30+ chains including Tron

The total gas sponsored by paymasters reached approximately $5.7 million, and while that number sounds small, it represents the subsidy that removed gas friction from 87% of all smart account transactions.

What Changed in the Last 12 Months

Three developments made the gasless stack production-ready:

1. Circle’s USDC Paymaster: An on-chain smart contract that lets users pay gas in USDC instead of ETH. No API keys, no developer account needed. Any ERC-4337 compatible wallet can use it. This removes the chicken-and-egg problem where users need ETH to do anything, including buying the USDC they would use to pay gas.

2. Gelato on Tron: Gelato extended its relay network to Tron with an innovative algorithmic energy-renting mechanism that cuts sponsored transaction costs by up to 80%. Tron processes more stablecoin volume than any other chain, and making those transactions gasless is a massive UX improvement for the emerging market users who dominate Tron’s user base.

3. Base’s Sub-Penny Economics: Base transactions cost approximately $0.007, and Coinbase’s Paymaster allows developers to sponsor gas entirely. When sponsoring gas costs less than a penny per transaction, the economics of gasless UX become trivial for any application with real usage.

The Remaining Challenges

Account abstraction is not a solved problem. Several challenges remain:

  • Cross-chain gas abstraction: Users still need to manage gas on different chains. Chain abstraction protocols are working on this, but seamless cross-chain gasless UX is not yet production-ready.
  • Paymaster sustainability: Who pays when the VC subsidies end? Applications need sustainable business models that can absorb gas costs.
  • Smart account fragmentation: Multiple smart account standards (Safe, Kernel, Biconomy, Coinbase) create fragmentation. ERC-7579 module standards are helping, but wallet portability remains limited.
  • Security: Paymasters introduce new attack vectors. More on this in a separate thread.

Despite these challenges, the tipping point has been reached. The default for new crypto applications should now be gasless-first, and the infrastructure exists to make that practical.

2

Brian, this is a great overview of the ecosystem. I want to share my experience actually building with these tools, because the developer experience has improved dramatically but there are still gotchas that the marketing materials do not mention.

Integrating ERC-4337 in a Real Application

I integrated smart account support into our DeFi protocol’s frontend last quarter. The process went roughly like this:

  1. Choose a smart account SDK: I went with Biconomy’s SDK because it had the best cross-chain support for our multi-chain deployment. Pimlico’s permissionless.js is more modular if you want lower-level control.

  2. Set up a bundler: We used Pimlico’s bundler because it supports the widest range of chains. The API is clean – you construct a UserOperation, send it to the bundler endpoint, and it handles gas estimation, bundling, and submission.

  3. Integrate a paymaster: This is where it gets interesting. We tried three approaches:

    • Coinbase Paymaster on Base: Free gas sponsorship with configurable per-user limits. Took about an hour to integrate on Base specifically.
    • Circle USDC Paymaster: Users pay gas in USDC. Integration required handling EIP-2612 permit signatures. Took about two hours.
    • Biconomy verifying paymaster: We sponsor gas and pay Biconomy monthly. Most flexible but adds a vendor dependency.

The total integration time was about two weeks for a production-ready implementation across three chains. A year ago, the same work would have taken two months.

EIP-7702 Changed Everything for Us

The biggest problem we had with ERC-4337 was user migration. Our existing users had EOA wallets with positions in our protocol. Asking them to create new smart wallets meant they had to migrate their positions – a multi-step process involving withdrawals, approvals, and re-deposits. Most users simply would not do it.

EIP-7702 solved this completely. Now our users can keep their existing MetaMask addresses and get smart account features (gasless transactions, batch operations, session keys) without any migration. The first time they interact with our updated frontend, their EOA temporarily delegates to our smart account implementation for that transaction. No new address, no position migration, no confusion.

What Still Frustrates Me

Gas estimation is unreliable. UserOperation gas estimation across different bundlers gives wildly different results. We have had transactions fail because the bundler underestimated gas, and we have had transactions overpay by 3-4x because the estimation was too conservative. There is no standard gas estimation algorithm.

Debugging is painful. When a UserOperation fails, the error messages from bundlers are often opaque. “AA25 invalid account nonce” does not tell me which nonce is wrong or why. The debugging tools are improving but they are not at the level of standard Ethereum transaction debugging.

Cross-chain UX is still fragmented. I can offer gasless on Base, Arbitrum, and Polygon separately, but I cannot offer a unified “gasless everywhere” experience. Each chain needs its own paymaster configuration, its own gas sponsorship budget, and its own UserOperation flow.

Despite these issues, I would never go back to building with raw EOA transactions. The UX improvement for users is too significant to ignore.

3

What Gasless UX Means for DeFi Protocol Design

Brian’s overview captures the infrastructure story well. Let me add the DeFi-specific implications, because gasless transactions change protocol economics in ways that are not immediately obvious.

Gas Costs Were a Hidden Tax on DeFi Users

I have modeled transaction costs across our yield strategies, and gas was a meaningful drag on returns for smaller positions. On Ethereum L1, a Uniswap swap costs $5-20 in gas. An Aave deposit-borrow-swap-deposit loop can cost $50-100. For a user with a $1,000 position doing weekly rebalancing, gas alone could consume 5-10% of annual yield.

On Base with paymaster-sponsored transactions, that same rebalancing loop costs the protocol $0.03-0.07 per user per week. At $0.07 per week, sponsoring gas for 10,000 weekly active users costs the protocol $700/week or about $36,000 per year. That is a trivial cost for any protocol generating meaningful revenue.

Batch Operations Change the DeFi UX Paradigm

With smart accounts, users can batch multiple DeFi operations into a single UserOperation. Instead of:

  1. Approve USDC → transaction 1 (gas)
  2. Swap USDC to WETH → transaction 2 (gas)
  3. Approve WETH on Aave → transaction 3 (gas)
  4. Deposit WETH on Aave → transaction 4 (gas)

…you get one atomic operation that does all four steps. This is not just a gas savings – it eliminates the failure modes where a user completes step 2 but their step 3 approval fails, leaving them with WETH they did not intend to hold.

For yield strategies that require multi-step execution, batching transforms the user experience from “click and wait four times” to “click once and you are done.”

The Revenue Model Question

Brian raised the sustainability question and it is the right one for DeFi protocols. The current model is mostly VC-subsidized gas sponsorship. But there are sustainable alternatives:

  1. Fee-inclusive sponsorship: The protocol charges a 0.1% fee on transactions and uses that revenue to cover gas. At Base’s sub-penny costs, even tiny fees generate surplus.
  2. Conditional sponsorship: Sponsor gas only for transactions above a minimum size, or only for users who hold the protocol’s governance token.
  3. Cross-subsidization: High-value transactions (large swaps, liquidations) subsidize gas for smaller users through protocol revenue.

The math works. If your protocol generates $1M in annual revenue and Base gas sponsorship costs $50K for 100,000 users, spending 5% of revenue on gas elimination is one of the highest-ROI UX investments you can make.

4

Gasless UX Is the Biggest User Acquisition Unlock in Crypto

Great analysis from everyone. Let me frame this from the business and product perspective, because as a founder who has struggled with user onboarding for three years, gasless transactions solve the single biggest conversion killer in crypto applications.

The Gas Problem Was Our #1 Drop-Off Point

We tracked our onboarding funnel obsessively. The largest drop-off was always at the “get ETH for gas” step. Users would connect their wallet, see that they needed ETH to do anything, and leave. Our data showed a 60-70% abandonment rate at this step for new users who came from non-crypto backgrounds.

With paymaster integration, that step disappears entirely. Users connect their wallet, and the first transaction just works. Our preliminary data shows a 3x improvement in onboarding completion rate since we implemented gasless on Base.

The CAC Math Finally Works

Customer acquisition cost in crypto has been absurdly high because of the onboarding friction. If you spend $5 to acquire a user through marketing and 65% of them abandon at the gas step, your effective CAC is $14.30 per converted user. Remove the gas step and your effective CAC drops to $6.25 (assuming a much lower 20% drop-off for other reasons).

At Base’s sub-penny transaction costs, sponsoring a user’s gas for their first 100 transactions costs less than $1. Compare that to the $9 per user you save in acquisition costs. The ROI on gas sponsorship is roughly 9x when measured against reduced abandonment.

Who Is Winning the Gasless Race

From what I see in the market, three players have disproportionate advantage:

  1. Coinbase/Base: They control the entire stack – the L2 (Base), the paymaster, the wallet (Coinbase Wallet with smart account), and the distribution (100M+ Coinbase users). No one else has this vertical integration.

  2. Circle: The USDC Paymaster is a horizontal play. It works on any chain, with any ERC-4337 wallet, and no API keys needed. Circle’s strategy is to make USDC the default gas token across all of crypto.

  3. Gelato: The most chain-agnostic infrastructure with 30+ chains including Tron. If you are building a multi-chain application, Gelato’s relay network is the closest thing to “gasless everywhere.”

For founders deciding where to build, my advice is: if you are going all-in on one chain, Base’s economics and Coinbase’s distribution make it the default choice. If you need multi-chain, Gelato or Biconomy. If you want users to pay gas in stablecoins rather than sponsoring it yourself, Circle’s USDC Paymaster.

The bottom line: gasless UX is no longer a nice-to-have feature. It is a competitive necessity for any consumer-facing crypto application.

5

The Investment Angle on Account Abstraction Infrastructure

Good thread. Let me add the market analysis because the 200 million smart account number has investment implications that the market has not fully priced in.

Who Captures Value in the Gasless Stack?

The gasless transaction stack has several layers, and the value capture varies:

Layer 1 (Settlement): Ethereum and L2s still earn fees on the underlying transactions. Gasless does not mean free – someone (the paymaster sponsor) still pays gas. ETH accrues value as the settlement layer.

Bundlers: Coinbase, Pimlico, Alchemy, and Biconomy run bundler infrastructure. Revenue comes from bundling fees (small margin on each UserOp) and paymaster-as-a-service subscriptions. Coinbase bundled 30M+ UserOps in 2024 – at even $0.001 per UserOp margin, that is $30K in revenue. Small today, but growing exponentially.

Paymaster Providers: Circle’s USDC Paymaster captures value through USDC adoption (more gas paid in USDC means more USDC demand). Gelato and Biconomy charge subscription or per-transaction fees for their managed paymaster services.

Smart Account Platforms: Safe, Biconomy, and Coinbase Smart Wallet compete for the account layer. Safe has 28.7M accounts but limited token value capture. Biconomy’s token (BICO) is the most direct play on account abstraction adoption.

The BICO and SAFE Token Thesis

BICO trades at a fraction of its 2022 highs despite 7x growth in smart account deployments. The disconnect is that Biconomy’s revenue model (paymaster and bundler fees) has not scaled proportionally to account deployments. Most of the 200M smart accounts are on Coinbase’s infrastructure, not Biconomy’s.

SAFE token has a stronger position in terms of accounts (28.7M) and TVL secured ($100B+), but the Safe ecosystem’s revenue capture is limited. The Bybit hack also created reputational headwinds.

My Positioning

I am bullish on the gasless UX thesis but cautious on individual token plays. The clearest beneficiaries are ETH (settlement demand), USDC (Circle Paymaster adoption), and the L2 tokens (Base/Coinbase stock, OP, ARB) that benefit from increased transaction volume driven by lower friction. The pure-play account abstraction tokens need to demonstrate sustainable revenue models before the market re-rates them.