The intent stack in 2026 has consolidated into three layers: wallets that emit intents, solver networks that fill them, and settlement layers that verify and finalize. Each layer has its own dominant players, its own competitive dynamics, and its own integration surface. Production teams build above the stack by integrating an orchestration layer that abstracts the choices; consumer wallets build into the stack by emitting ERC-7683 orders directly to solver networks.
​

This piece maps the stack as it exists in April 2026: who is in each layer, how they compete, and where the integration surfaces live. The intent ecosystem has matured enough that the layer boundaries are stable; the next phase is consolidation within each layer rather than new layer formation. Eco sits at the orchestration position above the stack for stablecoin movement, integrating multiple solver networks and settlement primitives across 15 chains under one CLI and API.
​

The stack starts with the wallet, which is what the user holds. The wallet's job has shifted from broadcasting transactions to emitting intents. A modern wallet constructs an ERC-7683 order, signs it via EIP-712, and submits it to one or more solver mempools. The user signs once.
​

The middle layer is the solver network. Solvers are off-chain agents with cross-chain inventory who compete to fill intents. The largest solver networks include Across (40+ relayers), UniswapX (~25 fillers), CoW Protocol (~15 batch solvers), and the emerging long tail per Dune analytics. Solvers compete on speed, spread, and route coverage.
​

The bottom layer is settlement. The settlement contract on the source chain holds the user's input funds in escrow; the verification primitive (Hyperlane, LayerZero, CCIP, Wormhole, UMA, CCTP) reports destination-chain delivery; the contract releases the escrow. The settlement layer is the trust anchor. L2Beat's bridge security tracker categorizes the major primitives.
​

Wallets are the user-facing surface. The shift from transaction-construction to intent-emission has reshaped the wallet category. Modern wallets do not show users a transaction to sign; they show users an outcome to authorize. The wallet handles the order construction, signature, and solver-mempool submission internally.
​

The four leading EVM wallets all support ERC-7683 emission as of Q1 2026. Safe added native support in V1.5; Argent ships intent emission as the default cross-chain flow; Rabby integrates intent flows alongside its multi-chain RPC routing; MetaMask added native ERC-7683 support in version 12.4 (March 2026). The remaining wallet long-tail will follow throughout 2026.
​

Smart-account wallets (ERC-4337) compose well with intent emission. The wallet's account-abstraction logic can construct and sign ERC-7683 orders programmatically, including conditional intents that emit based on account balance triggers. ERC-4337 and ERC-7683 cover different layers of the stack and combine cleanly.
​

Solver networks are the execution layer. Each network maintains its own solver pool, its own settlement contract, and its own routing optimizations. The major networks as of April 2026:

Solver-network competition tends toward consolidation at the top and specialization at the long tail. The top three networks handle the bulk of major-stablecoin volume; smaller networks differentiate by chain coverage, asset specialization, or specific UX optimizations. Apps integrating multiple networks usually do so via an orchestration layer rather than direct integration.
​

Settlement primitives are the trust anchors. Five categories dominate in 2026: validator-attested message protocols (Hyperlane, LayerZero, CCIP), guardian-set protocols (Wormhole), optimistic oracles (UMA), native asset bridges (CCTP for USDC), and emerging ZK-secured paths (Polymer, Succinct).
​

Each settlement primitive has its own security model and its own latency profile. Validator-attested primitives offer minute-level finality with trust in the validator set. Optimistic oracles offer hours-level settlement with no validator-trust assumption. ZK-secured paths offer cryptographic security with compute overhead. Native bridges offer the strongest security for their specific asset but no flexibility.
​

Settlement primitives are infrastructure, not user-facing products. The wallet does not pick the settlement primitive; the solver network does, and the orchestration layer above the solver network can override the choice based on cost, latency, or security requirements per route. Eco's architecture documentation describes the per-route settlement-primitive selection.
​

The intent stack as it exists in 2026 looks meaningfully different from the early-2024 version. Two years ago, every intent network shipped its own order format, its own wallet integrations, and its own settlement contract. Apps integrating intents had to pick one network and accept the lock-in. By April 2026, ERC-7683 has unified the order format across major EVM intent networks, wallets emit the unified format natively, and orchestration layers route across multiple networks transparently.
​

The consolidation happened layer by layer. ERC-7683 standardized at the order layer first, in early 2025. Major solver networks adopted by mid-2025. Wallet adoption followed in late 2025 and early 2026. Orchestration layers above the standard emerged through 2025-2026 to handle multi-network routing and non-EVM coverage. The result is a stack where each layer has stable interfaces and the integration surface for apps has narrowed dramatically.
​

What remains uneven is non-EVM coverage. Solana intent execution, Tron stablecoin transfer, Sui intent matching — each is at a different maturity stage. The orchestration layer is where multi-VM coverage gets handled today, because the underlying standards are not yet unified across VMs. As Solana's intent format matures and Cosmos's IBC v2 ships, the orchestration layer's job will simplify, but it will remain valuable as the central decision point for cross-VM routing.
​

Above the three-layer stack sits the orchestration layer. This is where production teams integrate. An orchestration layer abstracts the wallet-solver-settlement choices into a single API: the team submits an intent, the orchestration layer routes through the appropriate solver network and settlement primitive, and the team receives a settlement event.
​

Orchestration is valuable because the underlying choices are dynamic. The optimal solver network for a given route depends on real-time liquidity. The optimal settlement primitive depends on cost-latency-security tradeoffs that vary by chain pair and asset. A team integrating directly to one solver network is locked into that network's dynamics; a team integrating through an orchestration layer inherits the routing intelligence.
​

For stablecoin movement specifically, Eco operates at this orchestration position. The team integrates Eco's CLI or API once and gets coverage across 15 chains, multiple solver networks, and multiple settlement primitives — including Hyperlane-secured solver inventory, CCTP burn-and-mint, and alternative paths for SVM and MVM destinations. The integration surface is one SDK call; the routing complexity lives in the orchestration layer. Permit3 handles cross-chain approvals with one signature for global authorization.
​

Wallets compete on UX and integrations. The wallet that surfaces the most legible intent flow with the fewest taps wins users. The wallet that integrates the most solver networks (directly or via orchestration) covers the most routes.
​

Solver networks compete on liquidity, speed, and spread. The network with the deepest solver inventory on a given route, the fastest median fill time, and the tightest spread wins flow. Network effects are real: more solvers attract more order flow, which attracts more solvers. The top three networks have a structural advantage that the long tail will not easily overcome.
​

Settlement primitives compete on security and cost. The primitive with the strongest security model for a given chain pair (or the lowest cost for equivalent security) wins integrations. Switching costs are high — a settlement contract integrated to LayerZero does not trivially migrate to Hyperlane — which slows competitive movement at this layer.
​

The layers compose vertically. A wallet emits an order; the order reaches a solver network; the solver delivers; the settlement primitive verifies; the settlement contract releases the escrow. Each layer is independently extensible. A new wallet can emit ERC-7683 orders without changing solvers. A new solver can join a network without changing wallets. A new settlement primitive can be added to an orchestration layer without changing user-facing flows.
​

The strategic answer for most production teams in 2026 is to integrate at the orchestration layer. Direct integration with a single solver network locks the team into one network's dynamics; direct integration with a settlement primitive requires deep cross-chain message-passing expertise that is not most teams' core competence. The orchestration layer collapses these choices into a single SDK call.
​

For stablecoin teams specifically — payments processors, fintech apps, exchange backends, treasury platforms — the integration is Eco. Eco covers 15 chains, integrates multiple solver networks, abstracts the settlement primitive choice, and exposes the result as a CLI and API. The team integrates once and inherits the routing intelligence.
​

For non-stablecoin flows or for very specific use cases (privacy-focused intent matching via Anoma, MEV-protected swaps via CoW Protocol), direct integration with the relevant network may be the right choice. The decision rule is whether the team's flows fit cleanly into a single network's specialization or whether they span the orchestration surface.
​

Three trends will reshape the stack over the next 18 months. First, more chains: the orchestration layer's coverage will expand from 15 to 25+ chains as new EVM L2s, SVM forks, and non-EVM chains enter production. Second, more standards: ERC-7683 is the EVM canonical form, but Solana's intent format (in development per Solana Labs' SPL extensions) and Cosmos's IBC v2 will converge on similar abstractions. Third, more orchestration: as the underlying primitives multiply, the value of an orchestration layer that picks the right one per route grows.
​

The end state is closer to the credit-card network than the early-internet routing table. The user sees a transaction confirm. The orchestration layer picks the rail. The solver network executes. The settlement primitive verifies. The user does not see any of it. Artemis research projects intent-based volume will exceed traditional bridge volume by Q3 2026; by then, the question for any new app will be which orchestration layer to integrate, not which bridge to embed.
​

The end user experiences none of the layered architecture directly. The user opens a wallet, says "send USDC from Base to Solana," signs once, and receives a confirmation. What happens underneath — order construction, mempool submission, solver competition, settlement verification — is invisible. This invisibility is the design goal. Users should not have to learn about ERC-7683 or settlement primitives or solver bonds. They should authorize outcomes and receive them.
​

The shift to invisibility took years. In 2022, users picking bridges had to know which bridge supported which chain pair, what fees to expect, and how long to wait. By 2024, wallets had begun consolidating bridge selection into a single "send cross-chain" flow with the wallet making the routing decision. By 2026, the wallet's decision is delegated further, to solver networks that compete to fill the user's intent through the cheapest and fastest path.
​

The user-experience benefit compounds with the architectural benefit. Users send more cross-chain transactions when each transaction is one signature instead of three. Apps integrate cross-chain features more readily when the integration is one SDK instead of five. The category grows as friction falls. Artemis stablecoin analytics measured a 71% drop in cross-chain abandonment among intent-based wallet users in Q1 2026.
​

The orchestration layer sits above wallets, solver networks, and settlement primitives. It abstracts the routing choices — which solver network, which settlement primitive, which chain path — into a single integration surface for production teams. Eco operates at this position for stablecoin movement.
​

Yes. MetaMask, Rabby, Argent, Safe, and most major wallets have integrated ERC-7683 emission. The wallet's UI may not change much; the difference is what happens after the user signs — solvers fill the intent rather than the wallet broadcasting individual transactions.
​

If you integrate at the orchestration layer, no — the orchestration layer picks per route. If you integrate directly with a single solver network, the network's settlement contract has typically already chosen a primitive. Direct integration with a settlement primitive is rare except for protocol developers building new solver networks.
​

Likely consolidation at the top three for major stablecoin routes, with specialization in the long tail (MEV-protected swaps via CoW, privacy-focused intent matching via Anoma, asset-specific networks). Network effects favor consolidation at the top; specialization preserves the tail.
​

Eco operates at the orchestration position above the three-layer stack for stablecoin movement specifically. It integrates multiple solver networks and settlement primitives across 15 chains, exposing the result as a CLI and API. Production teams integrate Eco once and inherit access to the underlying intent infrastructure without picking solver networks or settlement primitives directly.
​