nordbench
DATA SOURCE · ENSO SHIELD ↗
Aggregator setup selector·built on Enso Shield data

Which aggregators does your setup need?

The same aggregator can be an excellent choice or a liability — it depends on how you integrate. Describe your setup below and this tool re-grades every aggregator against your reliability bar, using Enso Shield's measured data. The goal throughout: 99.9% execution reliability.

01

Describe your setup.

Start from a profile, then fine-tune. Everything below reacts live.

A shortcut — or build your own set by clicking rows in the table below. More independent sources means one can cover another's misses.
Simulate every quote against live state and rank by verified output — this filters out no-quotes, sim failures and overquoting before anything is sent. Without it, sim-failure becomes revert rate.
Target: 99.9% chance at least one quote arrives in time.
EXECUTION RELIABILITY
99%
99.9% target
98.0%99.0%99.9%100%

How it's computed

02

Every aggregator, graded for your setup.

Enso's measured values, unchanged. The colour bands shift with your setup: with more redundancy, no-quote and sim-failure relax; without simulation, sim-failure hard-pins to the revert bar; latency grades against your budget. Click any row to add or remove it from your set — the highlighted rows are what the reliability above is computed on. Auto-pick seeds a starting set; then it's yours to edit.

Aggregator Score1–10 · weighted Overquoteprice holds Quote→fill gapavg drift Sim failurereverts No-quoteavailability Median lat.speed p95 lat.tail
clears your bar marginal misses — would break 99.9% META = meta-aggregator (routes across other aggregators/bridges, not direct DEX liquidity)
03

Reliability funnel.

The set you selected above, traced through each gate. Every thin line is one source draining as quotes fail to return, arrive late, or fail simulation; the bold line is the combined result — redundancy recovers what individual sources lose.

Reliability funnel — where quotes drop, and how redundancy recovers
combined (≥1 survives) each thin line = one source draining through the gates · combined = 1 − ∏(all fail) at each step
04

Every aggregator's shape.

Six dimensions at once — further out is better on every axis. Your selected sources are drawn in colour; the rest of the field sits faint behind them. A round hexagon is an all-rounder; a spiky one trades strengths for weaknesses. Hover any shape for its name.

05

Latency against your budget.

A quote is only usable if it arrives in time. With one source you need the p95 inside budget; with several, the median is enough because the fastest of many arrives first. The line marks your current budget.

median p95 tail your budget bars past the line don't reliably arrive in time · chip = % of quotes arriving in time (availability × speed; = availability when budget is off)
06

Network coverage.

Execution quality is only half the picture — an aggregator also has to support the chains you ship on. Counts are refreshed server-side from verified machine-readable endpoints where providers expose one, normalized to mainnet swap networks only, and cached for one hour. Documented static values remain in place as fallbacks when a public discovery endpoint is unavailable or temporarily fails.

Nordstern live endpoint documented fallback mainnet swap networks only · loading server-side coverage…
07

Method & source.

All measurements are produced by Enso Shield — an independent benchmark that quotes every aggregator on the identical swap and re-simulates against live chain state (methodology ↗). nordbench changes none of the numbers; it only re-grades and re-plots them for a given integration setup.

The reliability model. Each source yields a usable quote if it arrives in time and — when simulating — passes simulation. Arrival is estimated from Enso's median and p95 via a log-normal fit. With independent sources, the request fails only if all miss, so reliability = 1 − ∏(1 − uᵢ). The green band for redundancy-covered failures (no-quote; sim-failure when simulating) is 0.001^(1/N): the per-source rate at which N sources still clear 99.9%. Without simulation, sim-failure becomes revert rate directly and redundancy can't mask it, so it hard-pins to 0.1%.

The honest limit. This assumes source failures are independent. Correlated failures — honeypots, transfer-tax tokens, genuinely illiquid routes — fail every source at once and set a floor no amount of redundancy clears. Reaching sub-0.1% in practice is two jobs: enough fast, honest, independent quotes (modelled here) and pre-trade filtering to remove intrinsically unexecutable routes (not in this dataset).

Disclosure: nordbench is published by Nordstern, itself one of the measured aggregators. Presentation is kept even-handed and the source is linked throughout so any figure can be verified directly against Enso.