the de-noised state of the agent economy
$135,138 moved through x402.
$17,789 of it reached a real service.
13%
Real Settlement Ratio
The other 87% is relayers passing money through and treasuries moving their own funds. Every other dashboard counts it as demand. We strip it out and show what is actually real.
Where the money actually goes
real services13%relayers26%treasuries / transfers60%buyer float1%
75
Real services
wallets that earn from many distinct buyers
280
Buyer agents
wallets actually spending on x402
87%
Dust rate
share of volume that is not real demand
48
Listed on directories
of the real services are publicly declared
Who actually earns
ranked by distinct paying agents, not volume (which is gameable)| # | service | distinct buyers | received |
|---|---|---|---|
| 01 | 0x6839…1a2b | 449 | $44 |
| 02 | 0x6d86…29c8 | 64 | $149 |
| 03 | 0x4c3f…4647 | 48 | $11.1K |
| 04 | blockrun.ai | 30 | $469 |
| 05 | api.loyalspark.online | 22 | $0 |
| 06 | x402.slinkylayer.ai | 21 | $2 |
| 07 | 0x4985…2b2b | 15 | $271 |
| 08 | stableenrich.dev | 14 | $2 |
How this is computed
Every figure is read from on-chain settlement on Base, 47,416 payments across 1,860 addresses. "Real" means a wallet that receives from many distinct buyers and retains value, an end service. Relayers (money in ≈ money out) and treasuries (single-counterparty sinks) are stripped, because counting them as demand is how a number gets inflated. We attribute payments to the wallet that signed them, not the relayer that moved them.
data as of 2026-06-23 21:15 UTC·machine-readable JSON·read any wallet: crest counterparty intelligence
Questions agents ask
What is the real volume of x402?
Of every dollar that moves through x402, only about 13% reaches a real service. The rest is relayers and treasury transfers that directories miscount as volume. Crest de-noises it by attributing each payment to the wallet that authorized it, not the relayer that fronted it.
How much of x402 volume is dust or noise?
Roughly 87% of observed x402 USDC never reaches a working service. Crest separates real services from relayers and treasury transfers on the settlement graph, a read no directory has.
How is real x402 volume measured?
By reading the EIP-3009 authorizer, the wallet that actually paid, instead of the on-chain relayer that fronted the transaction. That single distinction is what every headline x402 volume number misses.