Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

We tried to buy one call from 100 x402 sellers. Most delivered. Almost nobody is buying.

Description and use

Tested 100 x402 services: 74 of 85 paid calls returned genuine answers, 4 charged for nothing, 1 error after payment, 9 errors without payment. Most services work, but few buyers exist.

Problem

Assess reliability of x402 paid API endpoints by testing real payments and responses.

Outcome

Out of 85 paid calls, 74 (87%) returned genuine answers; 4 charged for nothing useful; 1 error after payment but charged; 9 errors after payment but not charged; 5 sellers not bought.

Workflow

  • Select 100 random eligible sellers with USDC $0.01 offer
  • Send dry-run unpaid request to read 402
  • Execute paid call with reference client capped at $0.01
  • Verify payment via USDC authorizationState
  • Judge answer quality with five-model jury
  • Record outcomes

Evidence

74 of the 85 paid calls (87%) returned a real answer. Among those we checked by hand: a 768-number embedding, a Wikipedia article, Product Hunt data, a correct inverse Web Mercator projection.

Source

A dry run first. One unpaid request to read the 402. Then the purchase, on exactly the terms the dry run saw.

Source

The reference x402 client, capped at $0.01 per call and allowed to sign only an EIP-3009 transferWithAuthorization.

Source

  • Record date: 2026-10-06T07:55:56.788Z