Skip to main content
We compared Route with Uniswap to see which offered more tokens for the same input. Route led on some trades and Uniswap led on others.
This is a September 6, 2026 snapshot of output before network gas. It does not show total swap value, live prices or completed trades. Route output was simulated; Uniswap output remained quoted. Quotes were collected at different blocks. Gas and approval costs are excluded and can change which service delivers better net value.

​
Before network gas

Uniswap is the 0 bps baseline for each input. Positive means Route offered more output; negative means Uniswap did. 1 basis point (bp) = 0.01%. These outputs already reflect applicable pool/provider pricing, so those costs have not been removed from the comparison.
Route versus Uniswap before network gas, in basis points. Mean Route advantages: minus 1.92, minus 0.617 and minus 0.00280 bps for 0.01, 0.1 and 1 ETH to USDG; plus 59.8, 74.4 and 103 bps for 0.01 ETH to PIPEDOG, PONS and CHUMP. Whiskers show both rounds.
Bars show the average of two rounds. The thin lines show the range between those results, not a confidence interval. Labels use three significant figures. Uniswap is normalized to zero separately for each input, which does not mean Uniswap returned zero tokens. Platform/interface fee: Route 0 bps · Uniswap Labs 0 bps. Uniswap’s published policy states zero Labs fees across its web app, wallet and Trading API. We do not add a historical interface surcharge. Source, checked September 6, 2026. Pool/provider pricing is already reflected in output; zero interface fees does not mean a cost-free swap.
Download all 12 before-gas observations in bps. The download keeps the unrounded figures.

​
What the results say

  • Core ETH/USDG output was close. The two-round averages slightly favored Uniswap before gas.
  • Route offered more output for PONS and CHUMP in this sample. PIPEDOG favored Route in one round and tied on output in the other.
  • These are output differences, not net savings. This chart does not establish which service is cheaper after execution costs.

​
What we tested

We requested 12 fixed inputs over two rounds, reversing their order in round two: 24 paired quotes in total. Each pair started concurrently, with at least five seconds between starts. Route used its deployed Vercel service with direct and Kyber routing enabled, 0x disabled, and no Route platform fee. Uniswap used the official Trading API’s BEST_PRICE setting without a protocol filter, Universal Router 2.1.1, a synthetic caller and 0.5% slippage. All 24 requests succeeded on each service. All Uniswap responses selected CLASSIC; no UniswapX fills were tested, and we did not confirm support for native ERC20-ETH. The chart covers the six native-ETH inputs, two observations each, for which we also built and checked Uniswap’s actual calldata. The other ERC20-input quotes are outside the chart. Successful quote-request median: 1.32 seconds for Route, 288 milliseconds for Uniswap. Route’s endpoint includes transaction assessment; Uniswap’s is quote-only. These timings measure different work, so they do not show which service completes a swap faster. Individual Route pool failures and rejected candidates remain in the raw evidence despite the successful top-level requests.

​
Limits and evidence

Uniswap’s output is a quote. We did not measure a change in the recipient’s token balance. An additional eth_simulateV1 attempt to recover output under matched block overrides failed; those failures are retained. The chart compares estimates, not completed swaps measured under identical conditions. Liquidity is not reserved, and the study does not establish broad token coverage or better results for every trade. Gas costs are not plotted on this page. No signatures, approvals or transactions were broadcast. Route’s contracts remain unaudited. Application source: af0ccaa. Raw quote SHA-256: c1b2c8385008d4fdf2edc8b9dff69aafb89d6bb845d030d73040edfa3a56c4d6. Uniswap quote and swap API