x402 Protocol Enabled

Built for AI Agents

Permissionless multi-model consensus intelligence. No accounts, no API keys, no registration. Your agent pays per request with USDC and gets structured, quality-scored answers instantly.

x402: Pay-Per-Request with USDC

The x402 protocol lets any AI agent pay for API calls with on-chain USDC on Base L2. Zero platform fees. No signup. Fully autonomous.

1

Send USDC

Transfer the exact amount in USDC on Base L2 to the LLM Consensus wallet. Fast mode costs $0.05, balanced $0.35, deep $4.50.

2

Encode Proof

Base64-encode a JSON object containing your transaction hash and sender wallet address:
{"txHash":"0x...","from":"0x..."}

3

Attach Header

Send your API request with the header X-Payment: <base64>. The server verifies on-chain and processes your prompt.

Wallet Address (Base L2)
0xB0FdC6030B9f30652e8B221B8090d443Dd3C6381
Network: Base (Coinbase L2) · Token: USDC · Fee: 0%
Mode Price Models
fast $0.05 GPT-5.4 Mini, Claude Haiku 4.5, Gemini 2.5 Flash
balanced $0.35 GPT-5.4, Claude Sonnet 4.6, Gemini 3 Pro
deep $4.50 GPT-5.4, Claude Opus 4.6, Gemini 3.1 Pro, Mistral Large 2, Llama 3.3 70B
Read the full x402 specification

Three Files Your Agent Needs

Standard discovery files so AI agents can find, understand, and integrate with LLM Consensus automatically.

TXT

/llms.txt

Human- and machine-readable service overview. Endpoints, modes, pricing, and capabilities in plain text.

# LLM Consensus API > One API call orchestrates multiple frontier LLMs into superior answers. Base URL: https://llmconsensus.io/v1 Auth: X-API-Key or X-Payment (x402) POST /v1/orchestrate POST /v1/batch
View full file
JSON

/.well-known/ai-plugin.json

OpenAI plugin manifest. Allows ChatGPT and compatible agents to discover and use this API natively.

{ "schema_version": "v1", "name_for_model": "llm_consensus", "api": { "type": "openapi", "url": "https://llmconsensus.io /openapi.json" } }
View manifest
API

/openapi.json

Full OpenAPI 3.1 specification. Every endpoint, schema, parameter, and response type, machine-parseable.

{ "openapi": "3.1.0", "info": { "title": "LLM Consensus API", "version": "1.0.0" }, "paths": { "/v1/orchestrate": { ... }, "/v1/batch": { ... } } }
View spec

What Your Agent Receives Without Auth

Call the API without credentials and you get a spec-compliant 402 Payment Required response with everything your agent needs to pay.

402
Payment
Required
application/json
{
  "error": "Payment Required",
  "message": "This endpoint requires payment. Use x402 or an API key.",
  "accepts": [
    {
      "scheme": "x402",
      "network": "base",
      "token": "USDC",
      "address": "0xB0FdC6030B9f30652e8B221B8090d443Dd3C6381",
      "pricing": {
        "fast": "0.05",
        "balanced": "0.35",
        "deep": "4.50"
      },
      "spec": "https://x402.org"
    },
    {
      "scheme": "api_key",
      "header": "X-API-Key",
      "register": "https://llmconsensus.io/v1/auth/register"
    }
  ]
}

Python: AI Agent with x402

A complete example of an AI agent paying for and using the LLM Consensus API with the x402 protocol.

agent.py
import json, base64, requests
from web3 import Web3

# --- Config ---
BASE_URL     = "https://llmconsensus.io/v1"
WALLET_TO    = "0xB0FdC6030B9f30652e8B221B8090d443Dd3C6381"
USDC_BASE    = "0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913"  # USDC on Base
MODE         = "balanced"
PRICE_USDC   = 0.35

# --- Step 1: Send USDC on Base L2 ---
w3 = Web3(Web3.HTTPProvider("https://mainnet.base.org"))
usdc = w3.eth.contract(address=USDC_BASE, abi=ERC20_ABI)
tx = usdc.functions.transfer(
    WALLET_TO,
    int(PRICE_USDC * 1e6)  # USDC has 6 decimals
).build_and_send(account)
receipt = w3.eth.wait_for_transaction_receipt(tx)

# --- Step 2: Encode proof as base64 ---
proof = json.dumps({
    "txHash": receipt.transactionHash.hex(),
    "from":   account.address
})
payment_header = base64.b64encode(proof.encode()).decode()

# --- Step 3: Call the API with X-Payment header ---
response = requests.post(
    f"{BASE_URL}/orchestrate",
    headers={"X-Payment": payment_header},
    json={
        "prompt": "Compare quantum error correction approaches",
        "mode":   MODE,
        "domain": "auto"
    }
)

result = response.json()
print(f"Consensus answer (score: {result['quality_score']}):")
print(result["response"])

Also for Human Developers

Prefer API keys and a dashboard? Register for an account, get 5 free credits, and pay with Stripe or crypto top-up. Same API, same quality, traditional auth flow.

API Key + Stripe USDC Top-Up (NOWPayments) 5 Free Credits on Signup

Create Account