# What is a DEX aggregator? A DEX aggregator is software that simultaneously queries multiple decentralized exchanges to determine the best of the answers. It compares potential payouts, subtracts any associated fees and gas costs, and executes the trade through the most favorable route, which can sometimes involve several pools. A DEX aggregator does not operate its own pools and never takes custody of funds. The scale of this routing capability is now quantifiable. DEX aggregators routed about $78.6 billion in transactions over 30 days to 12 August 2026, representing an increase of 20.6% compared to the previous 30 days, spanning 174 chains ([DefiLlama](https://api.llama.fi/overview/aggregators)). During the same period, all DEXes tracked by DefiLlama recorded a total of about $194.4 billion, involving 1,303 protocols across 287 chains ([DefiLlama](https://api.llama.fi/overview/dexs)). By dividing these figures, aggregator flow accounts for roughly 40% of DEX volume, based on arithmetic between two of DefiLlama's datasets, rather than a published figure. Since routed trades settle on the pools to which they were routed, this ratio indicates how much on-chain trading now occurs through a router rather than manually. ## How does a DEX aggregator work? There are four key steps from entering a trade amount to signing a transaction. First, the request is sent to every source covered by the aggregator for that specific chain. This process happens simultaneously, so a slow exchange only delays its own response without affecting others. Next, each response is priced based on the trade size, not a standard rate. Automated market makers adjust pricing along a curve, reducing the payout per token as the trade size increases relative to the pool's depth. Hence, a quote is tailored to a specific amount, and the ranking of two pools might change between a small and a large trade. The responses are then ranked based on the net amount received in the wallet, accounting for gas costs. A multi-pool path might offer a higher quoted amount but still result in a lower net payout after costs, illustrating why solely selecting the highest quote can lead to a suboptimal trade. Finally, the chosen route is executed in a single transaction. If any part of a split transaction fails, the entire transaction reverts, leaving the trader with their original funds. Splitting distinguishes routing from simple comparison shopping. Sending an entire order to the best-priced pool can adversely affect its price to the point where using three adequately priced pools yields a better average result. This calculation is fundamental to the aggregator's function. ## DEX vs DEX aggregator: what is the difference? A DEX provides liquidity. Its pools act as the counterparty, charging fees on trades and setting prices based on their reserves. Uniswap and PulseX are examples of such DEXes. In contrast, a DEX aggregator does not hold liquidity. It lacks pools, an order book, and inventory, and cannot create liquidity independently. Its role is to compare and create paths through existing pools. A good aggregator is non-custodial, meaning swaps settle directly from the trader's wallet to a router contract in a single transaction, without necessitating deposits or accounts. The practical difference is evident in the information each provides. A DEX offers its own prices, requiring traders to manually verify competitiveness. An aggregator already performs this comparison, making the relevant question whether it transparently displays its process. ## Do aggregators get better prices? Typically, for a specific pair and size, an aggregator can outperform any single exchange it queries. However, this is not guaranteed or optimal in every instance. The strongest public measure of the gap is from an audit of 2.98 million WETH-USDC swaps on Ethereum. This audit compared actual trader routes to an optimal benchmark, finding an average shortfall of 2.02 basis points per trade, totaling approximately $24 million ([Xi and Moallemi, arXiv:2607.20762](https://arxiv.org/abs/2607.20762)). As a preprint rather than peer-reviewed work, it focuses on one pair on one chain, emphasizing detail over headline figures. Percentage losses skew towards smaller trades, with sandwich attacks contributing significantly to perceived poor routing. Another study on solver-based exchanges found improved execution relative to routing directly through Uniswap V2 or V3, but only within specific trade-size ranges. This was clearer for the liquid USDC-WETH pair than for less common pairs ([Yuminaga, Chen and Sui, arXiv:2503.00738](https://arxiv.org/abs/2503.00738)). Also a preprint, the study highlights that routing is beneficial where alternatives exist. > A figure that should not influence decision-making is the number of liquidity sources an aggregator claims. 0x lists three different coverage figures: 370+ on its homepage, 150+ in its developer documentation, and 130+ on its swap API product page. These counts are marketing tools. What matters is whether quotes account for fees and gas, and whether unsuccessful routes remain visible. ## Where did this mechanism come from? The equities market pioneered this mechanism, driven by legal mandates rather than competition. The SEC implemented Regulation NMS on 9 June 2005, with Rule 611, the Order Protection Rule, requiring trading centers to keep written policies reasonably designed to prevent executions at prices worse than another market's displayed quote ([SEC Release No. 34-51808](https://www.sec.gov/files/rules/final/34-51808.pdf)). Smart order routing was developed to meet this requirement. DeFi adopted a similar mechanism independently, addressing greater on-chain fragmentation and focusing on better transaction fills. [Smart order routing](https://pulseswap.io/learn/smart-order-routing) provides insights into this history and its implications for quotes. ## What is an aggregator of aggregators? This concept operates one level higher. Rather than querying only DEXes for pricing, the router also includes other aggregators, ranking all responses together. The improvement is mathematical, not magical. Including an aggregator's response allows matching its performance when it offers the best outcome and surpassing it when direct pool routes are superior, as it evaluates all available options plus an additional one. PulseSwap employs this structure across 17 blockchains. On PulseChain, quotes from 11 DEXes and aggregators, the routing partner Piteas and PulseSwap's own router among them, are requested simultaneously and ranked after gas. On other supported chains, quotes are sourced from commercial aggregator partners. [Aggregator of aggregators](https://pulseswap.io/learn/aggregator-of-aggregators) explains the benefits and limitations of this approach. ## Sources - DefiLlama, DEX aggregator volume, read 12 August 2026. - DefiLlama, DEX volume across all tracked protocols, read 12 August 2026. - Weiye Xi and Ciamac C. Moallemi, "Quantifying Sub-Optimality in Routing for Automated Market Makers", arXiv:2607.20762, submitted 22 July 2026. Preprint, not peer-reviewed. - Yuki Yuminaga, Dex Chen and Danning Sui, "Execution Welfare Across Solver-based DEXes", arXiv:2503.00738, March 2025. Preprint, not peer-reviewed. - 0x, published liquidity-source counts across its homepage, developer documentation and swap API product page, read 12 August 2026. - SEC, Regulation NMS adopting release, Release No. 34-51808, 9 June 2005 (Rule 611, the Order Protection Rule). --- Canonical HTML page: