Last month we launched the $10/month GoldRush Vibe Coding Plan - the lowest-friction way to start building with onchain data. But we kept asking ourselves: what if there was no plan at all? No account, no credit card, no billing page. Just pay for exactly what you use, directly from a wallet.
Today we’re launching GoldRush x402 - a payment-gated proxy that lets you access the full GoldRush blockchain data API by paying per request using the x402 protocol. It’s live on Base Sepolia testnet now, with mainnet coming soon.
The real unlock isn’t just cheaper API calls - it’s what happens when AI agents can pay for data themselves.
Picture a portfolio monitoring agent. You give it a wallet address and a budget. It takes it from there:
import { HTTPClient } from "@x402/core";
import { ExactEvmScheme } from "@x402/evm";
const client = new HTTPClient({
scheme: new ExactEvmScheme({
network: "eip155:84532",
privateKey: process.env.AGENT_WALLET_KEY,
}),
});
const BASE = "https://x402.goldrush.dev/v1";
// Step 1: Agent discovers what endpoints exist (free)
const { endpoints } = await client.get(`${BASE}/x402/endpoints`);
// Step 2: Agent checks token balances across chains
const chains = ["eth-mainnet", "base-mainnet", "arbitrum-mainnet"];
const balances = await Promise.all(
chains.map((chain) =>
client.get(`${BASE}/${chain}/address/${walletAddress}/balances_v2/`)
)
);
// Step 3: Agent spots a large new position, pulls transaction history
const largeHoldings = balances
.flatMap((b) => b.data.items)
.filter((token) => parseFloat(token.quote) > 10000);
for (const token of largeHoldings) {
// Agent selects tier based on its budget - start small, upgrade if needed
const txns = await client.get(
`${BASE}/eth-mainnet/address/${walletAddress}/transactions_v3/?tier=small`
);
// Agent analyzes transactions, flags unusual activity, reports back
}No human approved those API calls. No one provisioned credentials. The agent discovered endpoints at runtime, evaluated pricing from response headers, decided which calls were worth making, and paid from its own wallet. Each agent gets its own wallet and its own rate limits - no shared API key to rotate across a fleet of agents.
HTTP status code 402 Payment Required has existed since 1997. It was reserved for “future use” - a placeholder for a world where the internet had native payments. Nearly three decades later, the x402 protocol finally makes it real.
The protocol has already processed over $600 million in payment volume across AI platforms, cloud providers, and blockchain ecosystems - with companies like Cloudflare, Google Cloud, and Anthropic integrating it into their infrastructure.
When you call a GoldRush x402 endpoint without payment, the server responds with 402 Payment Required along with payment instructions. Your client pays using stablecoins on Base, retries the request with proof of payment, and gets the data. The entire flow happens in a single request-response cycle.
No accounts. No API keys. No invoices. No minimums.
GoldRush x402 is a transparent reverse proxy sitting in front of api.covalenthq.com. It exposes the same paths and returns the same responses - the only difference is how you authenticate. Instead of an API key in a header, you attach a micropayment.
Your App ──── x402 payment ────▶ GoldRush x402 Proxy ────▶ api.covalenthq.com
│
validates request
verifies payment
caches response
returns data
The proxy validates your request before charging you. If the chain doesn’t exist, the address is malformed, or the endpoint doesn’t support that network - you get a clear error and pay nothing.
The discovery API is free and requires no payment. Use it to explore what’s available:
# List all endpoints with pricing
curl https://x402.goldrush.dev/v1/x402/endpoints | jq
# Search for specific functionality
curl https://x402.goldrush.dev/v1/x402/search?q=balance | jq
# Get details for a specific endpoint
curl https://x402.goldrush.dev/v1/x402/endpoints/get-token-balances-for-address | jqEvery endpoint returns its credit rate, pricing model, supported chains, and x402 payment instructions - everything a client (or an AI agent) needs to decide whether to make the call.
Most endpoints have a fixed cost per request. Token balances, NFT holdings, block details - one price, one call, one response.
Endpoints that return variable-length data (transaction histories, event logs) use tiered pricing. You select a tier upfront based on how much data you expect:
If the response contains fewer items than your tier covers, you overpaid slightly but get the data immediately. If it contains more, you get a 402 telling you exactly which tier you need. This keeps pricing predictable while accounting for variable response sizes.
60+ published endpoints covering balances, NFTs, transactions, blocks, DEX data, pricing, security, and cross-chain operations
100+ blockchain networks including Ethereum, Base, Arbitrum, Optimism, Polygon, Solana, Bitcoin, and dozens more
Response caching with intelligent TTLs (30s for balances, 5m for pricing data) - cached responses cost less
Per-wallet rate limiting at 100 requests per minute by default
Request validation before payment - never pay for a request that would fail
GoldRush x402 is live on Base Sepolia testnet. You can start building and testing today with zero cost - testnet stablecoins are free. This is the time to integrate x402 into your app or agent workflow, validate your use case, and be ready when we flip the switch to mainnet.
You can track GoldRush x402’s transaction volume and payment activity on x402scan alongside the rest of the x402 ecosystem.
We’re actively working on:
Base mainnet launch - real-value payments, same API, same integration
Credit system for overpayments on linear-priced endpoints
WebSocket streaming support with per-message micropayments
Open-sourcing the proxy so you can run your own instance or contribute
Distributed caching and rate limiting for global scale
GoldRush x402 sits alongside the $10/month Vibe Coding Plan as another way to get started with onchain data - but for a different kind of builder. The Vibe Coding Plan is ideal when you want a predictable monthly cost and an API key you can drop into any project. x402 is for when you want true pay-as-you-go with no commitment, or when you’re building for AI agents that need to pay for data autonomously.
Both paths get you the same data, the same endpoints, the same quality. Pick the one that fits how you build.