The programmable compute + data market for AI agents. Discover, pay for, and use compute in a single request — settled machine-to-machine over HTTP 402 in USDC, no account or API key.

x402GPU / CPU computeLLM inferencerent out your GPUBase · Polygon · Arbitrum · Solanano signupMCP

Compute & inference agent-native

Endpoint	Price	What it does
POST /v1/compute	per-second	Rent metered GPU/CPU for a container job. Pay per second in one x402 flow; scheduling prefers nodes with your model already warm; returns a signed, verifiable execution receipt.
POST /v1/infer	per-token	Open-model LLM inference — Llama 3.1 8B/70B, Qwen 2.5, Mistral. Prepay prompt + max_tokens, billed on actual usage with the rest credited back; signed usage receipt. From $0.05/1M input tokens.

Live endpoints and prices always appear in /v1 and /.well-known/x402.json.

Rent out your GPU earn per call

Running an OpenAI-compatible endpoint (vLLM, Ollama, TGI, LM Studio, llama.cpp)? Connect your wallet + endpoint and earn — agents discover you and pay per call, settled to your wallet. No dashboard, no invoicing; the payment is the contract.

One command to list. It health-checks your endpoint, signs the ownership proof locally with your wallet (EIP-191 — the key never leaves your machine), and registers you:

npm i viem
curl -O https://gate402.app/register-provider.mjs
node register-provider.mjs \
  --endpoint https://your-host/v1 \
  --model llama-3.1-8b \
  --price 0.02 \
  --key 0xYOUR_WALLET_PRIVATE_KEY

Your endpoint must be on a public HTTPS URL that answers GET /v1/models — a tunnel or reverse proxy in front of a home box is fine; localhost and LAN addresses are rejected. Minimum price is ~$0.0134/call (below that the fee can't cover payout gas).

Step	What
1 · list	POST /v1/providers/register — sign a message with your payout wallet to prove ownership, then list your endpoint, model, and price. (The script above does this for you.)
2 · get found	Agents discover you at GET /v1/providers and pay via POST /v1/market/infer. Providers are ranked by reliability band, with price deciding inside the band — and your first calls count as a trial run, so a new listing can win on price from day one.
3 · earn	You keep 85% of every successful call, credited to your wallet on delivery (escrow-on-success — a failed call never charges the buyer or pays you). gate402 takes 15%. Payouts sweep hourly in USDC on Base once your balance clears $0.10.
4 · stay in control	POST /v1/providers/:id/update with { priceUsdc?, endpointUrl?, active?, ts, signature } — change your price, move your endpoint, or delist with active:false. Signed by the same payout wallet; you're never locked in.

Finance endpoints for trading agents

Endpoint	Price (USDC)	What it does
POST /v1/onchain	$0.01	Base wallet/token intel: native + ERC-20 balances, EOA/contract, tx count, token metadata.
POST /v1/dex	$0.01	Live DEX price / liquidity / 24h volume across all Base DEXs, with cross-DEX arbitrage spread + timestamps.
POST /v1/news	$0.02	Recent news headlines + heuristic bull/bear sentiment for a ticker or topic.
POST /v1/edgar	$0.02	Latest SEC filings (10-K/10-Q/8-K) for a US ticker, with document links.

General agent endpoints

Endpoint	Price (USDC)	What it does
POST /v1/proxy	$0.002	URL → clean LLM-ready Markdown (renders JS).
POST /v1/proxy/stealth	$0.05	Cloudflare/anti-bot fetch via residential proxies + CAPTCHA solving → clean Markdown.
POST /v1/minify	$0.005 / 10k tok	Compress text, cut upstream LLM token bills ~40%.
POST /v1/dedup	$0.001 hit / $0.003 miss	Semantic vector cache, sub-10ms in-memory hits.

Use it from Claude, Cursor, Cline …

Everything ships as an MCP server (17 tools). Drop it into any MCP client — a free-credit key is claimed automatically on first use, no wallet or signup needed:

{
  "mcpServers": {
    "gate402": { "command": "npx", "args": ["-y", "gate402-mcp"] }
  }
}

Compute & inference: gate402_infer · gate402_compute.
Finance: gate402_onchain · gate402_dex · gate402_token_risk · gate402_momentum · gate402_best_swap · gate402_launches · gate402_news · gate402_edgar.
Web tools: gate402_scrape · gate402_scrape_stealth · gate402_minify · gate402_dedup.
Free (local, no key): gate402_token_count · gate402_html_to_md · gate402_json_repair.

npm: gate402-mcp · source: github.com/pgalyen1987/gate402-mcp

Independent trust scores

Live reliability grades from x402.fuchss.app — uptime, spec compliance, and settlement track record, refreshed continuously.

Machine-readable

Discovery manifest: /.well-known/x402.json
Full catalog + input schemas: /v1

$1 human utilities

File, data, and crypto micro-tools with card checkout for humans and x402 for agents — /tools. Each call is $1.

Quickstart

Every call returns HTTP 402 with payment requirements. Pay with an x402 client, or present an X-API-Key.

1 — see the price (any HTTP client):

curl -s -X POST https://gate402.app/v1/minify \
  -H 'content-type: application/json' \
  -d '{"text":"strip the filler from this verbose passage..."}'
# -> 402 { "x402Version": 2, "accepts": [ ...Base/Polygon/Arbitrum/Solana... ], ... }

2 — pay with x402 (e.g. x402-fetch or any agent x402 client), or with an API key:

curl -s -X POST https://gate402.app/v1/minify \
  -H 'content-type: application/json' \
  -H 'X-API-Key: sk_live_...' \
  -d '{"text":"..."}'

No wallet? Claim a free-credit key — POST /v1/free-key — then send it as X-API-Key. Top up at /ops/billing/checkout when it runs out.