Traditional blockchain transactions are imperative: you specify every detail of what must happen — which contract to call, which tokens to trade on which DEX, at what price, via which route. If these instructions produce a bad result (due to slippage, MEV, or a bad route), that’s the user’s problem. Intents are declarative: you specify the outcome you want, and specialized market participants (“solvers”) compete to figure out the best way to achieve it. The solver who can give you the best price wins the order. This paradigm shift enables gasless transactions, protection from sandwich attacks, cross-chain swaps, and a general-purpose auction mechanism for on-chain actions — and represents one of Ethereum’s most active research and deployment areas in 2024-2025.
Intents vs. Traditional Transactions
Traditional (imperative) swap:
“`
User specifies:
- Exact contract: Uniswap v3 Router
- Exact path: USDC → ETH via 0.3% pool
- Exact slippage: max 0.5%
- Gas: user pays in ETH
- Execute on: Ethereum mainnet
“`
The transaction does exactly what was specified — even if there was a better price on another DEX or another routing path available.
Intent-based swap:
“`
User declares:
- Give: 1000 USDC
- Receive: at least 0.39 ETH
- Deadline: within 5 minutes
- Signer: user’s wallet
“`
Solvers compete: one solver might route through Uniswap, another through CoW Protocol, another through direct OTC matching. The solver that provides the best execution (more than 0.39 ETH) wins and earns the improvement as profit.
The Solver Ecosystem
What solvers do:
- Monitor an order pool for signed user intents
- Simulate execution paths: DEX routing, L2 transfers, OTC matching
- Determine if they can profitably fill the order (give user X, keep remainder)
- Submit the winning fulfillment on-chain
Solver competition:
Multiple solvers compete for the same order. The solver offering the best outcome to the user (or equivalently: the one with the most efficient execution path) will be selected.
Solver profit source:
- Better routing: solver finds 0.401 ETH from routing, gives user 0.390 ETH, keeps 0.011 ETH
- Cross-chain arbitrage: solver takes user USDC on Ethereum, gives ETH on Arbitrum, profits from price differential
- Private liquidity: solver has private inventory or OTC access with better prices
Major Intent Protocols
The following sections cover this in detail.
UniswapX
- Users sign “Dutch orders” (prices slide down over time if not immediately filled)
- Fillers (solvers) compete to fill orders before preferred price expires
- Cross-chain swaps: same intent mechanism works across chains (UniswapX Cross-Chain)
- Gasless: gas cost folded into the order outcome difference (filler pays gas, recovers from spread)
Architecture:
- User signs an intent order (off-chain)
- Order posted to UniswapX order pool
- Fillers monitor order pool, compute profitable fills
- Winning filler submits on-chain transaction within deadline
CoW Protocol
When direct CoW matches aren’t available, CoW routes through DEXes but still uses a batch auction:
- User orders collected over a period (e.g., 30 seconds)
- Solvers compete to find the best batch execution (some orders matched directly, rest DEX-routed)
- Protects from sandwich attacks (no real-time ordering, batch settled simultaneously)
MEV protection: CoW Protocol is specifically designed to prevent MEV. The batch settlement model means frontrunning individual transactions is economically unviable.
1inch Fusion
- Market makers (“resolvers”) compete to fill orders
- Positive slippage shared with user
ACROSS Protocol (Bridge Intents)
- Relayers fill instantly on destination chain with their own liquidity
- Repaid from source chain bridging (user experience: near-instant cross-chain)
Account Abstraction (AA) and Intents
ERC-4337 (Account Abstraction) on Ethereum is complementary to intents:
- AA allows programmable validation of transactions (any condition can authorize a tx, not just ECDSA signature)
- Combines with intents: user’s smart account authorizes the solver’s fulfillment if it meets the declared intent
- Gasless user experience possible: solver/paymaster covers gas
Entrypoint + bundlers:
AA bundlers aggregate user operations (UserOps) and submit them to the entrypoint contract. Combined with intent solvers, end-to-end UX becomes: user signs intent → solver finds execution → bundler submits → user never touches gas.
Intent Use Cases Beyond Swaps
Cross-chain actions:
“Bridge 1 ETH from Ethereum to Arbitrum and deposit into Aave” — single intent, solver handles bridging + DeFi interaction
Limit orders:
“Buy 1 ETH if price reaches $2,400 at any point in the next 7 days” — keeper bots monitor and fill when condition met
Subscription payments:
“Allow Alice to withdraw 100 USDC on the 1st of every month” — time-conditional intents
DeFi automation:
“Compound my staking rewards automatically” — ongoing delegated execution
Intents and MEV
One of intents’ key claimed benefits is MEV protection:
Traditional transaction MEV exposure:
- User broadcasts swap transaction to mempool
- MEV bot spots it, sandwiches it (buy before, sell after)
- User gets worse price
Intent MEV protection:
- User signs an intent (off-chain, private until filled)
- Solver fills the intent — solver owns the execution path
- Public mempool never sees the intent directly
- Solver’s filling transaction might still be MEV’d, but that’s solver’s problem, not user’s
How to Use Intent-Based Protocols
Simplest path: just use UniswapX or CoW Protocol — they’re integrated into Uniswap and CoW Swap UIs respectively. No awareness of “intents” required.
- Go to app.uniswap.org or swap.cow.fi
- Select tokens and amount
- Sign the intent order (no gas at signing time usually)
- Wait for fill (<30s typically)
- Confirm on-chain result
To access: Get tokens via , use EVM wallet (MetaMask, etc.), store long-term at
Social Media Sentiment
Intents are one of Ethereum’s most hyped DeFi UX innovations. The concept has real merit — CoW Protocol’s batch auction model demonstrably reduces MEV compared to traditional orderflow, UniswapX provides better pricing for many swaps due to solver competition, and the cross-chain UX improvements are genuine. The key debates: 1) Whether intent-based systems are truly better for users, or just redistribute the “take” from MEV bots to solvers (some argue solvers are just more sophisticated MEV bots); 2) Whether solver centralization (a few large solvers dominate most order flow) creates new trust assumptions; 3) Whether the complexity of intent-based systems creates new attack surfaces. The Ethereum research community is actively exploring “programmable intents” as the next frontier — general-purpose agent-based execution replacing all imperative transaction models.
Last updated: 2026-04
Related Terms
Sources
Susskind, A., & Evan, N. (2023). Intents: A New Transaction Paradigm. Essential Ethereum.
Angeris, G., Chitra, T., Evans, A., & Roughgarden, T. (2023). Optimal Routing for Constant Function Market Makers. arXiv:2204.05238.
Daian, P., Goldfeder, S., Kell, T., Li, Y., Zhao, X., Bentov, I., Breidenbach, L., & Juels, A. (2020). Flash Boys 2.0: Frontrunning in Decentralized Exchanges, Miner Extractable Value, and Consensus Instability. IEEE Symposium on Security and Privacy.
Klages-Mundt, A., Canidio, A., & Roughgarden, T. (2024). Credible Decentralized Exchange Design via Verifiable Sequencing Rules. arXiv:2209.15569.
Roughgarden, T. (2021). Transaction Fee Mechanism Design. ACM SIGecom Exchanges.