LLM inference

LLM inference proxy - send an OpenAI-format chat/completions request and get a response from GPT-4o-mini. Send POST /api/llm with the required fields model and messages and pay $0.010 per call over x402 or MPP (there is no free tier). It returns a JSON object with model, provider, usage and choices.

Supports vision (up to 2 image URLs, low detail) and structured output (response_format: json_object or json_schema). No API key needed; pay per call via x402. Input capped at 16k chars, output at 4096 tokens.

Category: AI & compute · Tags: llm ai inference chat proxy openai gpt-4o-mini vision

TRY IN PLAYGROUND →

Parameters

Example request

curl -i -X POST https://agent402.tools/api/llm \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Say hello in one sentence."}],"max_tokens":64}'

Without payment this returns HTTP 402 Payment Required with the exact price for llm; any x402 v2 or MPP client pays it and retries.

Example response

{
  "model": "gpt-4o-mini",
  "provider": "openai",
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 8,
    "total_tokens": 20
  },
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you today?"
      },
      "finish_reason": "stop"
    }
  ]
}

From an MCP client

catalog.call {
  "slug": "llm",
  "params": {
    "model": "gpt-4o-mini",
    "messages": [
      {
        "role": "user",
        "content": "Say hello in one sentence."
      }
    ],
    "max_tokens": 64
  }
}

The hosted connector at https://agent402.tools/mcp needs a payment for llm; the stdio package pays it from a wallet or from AGENT402_CREDITS_KEY. Local install: npx -y agent402-mcp.

Errors and behavior

• model and messages are required. An input the tool rejects returns an HTTP 4xx whose body carries error, tool, expected, required and example, so the caller can correct it.
• A paid call that ends in any status of 400 or above is not charged over x402, MPP or a prepaid credits key: settlement is cancelled when the tool fails. The exception is a Tempo push credential, a transfer the buyer sent before the call: it settles before the tool runs, so if the tool then fails the payment is recorded as a refund owed to the paying wallet.
• Wallet-only: this tool runs a model, so it has no proof-of-work tier. A prepaid card-credits key issued earlier (Authorization: Bearer a402_...) also pays it.
• Model-backed: the answer is generated by a model, so the same input can produce different wording.
• A GET or HEAD to /api/llm returns the same 402 quote, so the price can be read without a body.
• An Idempotency-Key header makes a retried paid call replay the first 200 instead of charging again (an answer larger than 1 MB is not replayed).

Paid call (JavaScript agent)

import { wrapFetchWithPayment } from "@x402/fetch";
import { x402Client } from "@x402/core/client";
import { registerExactEvmScheme } from "@x402/evm/exact/client";
import { privateKeyToAccount } from "viem/accounts";

const client = new x402Client();
client.setSpendControls?.(false); // keep your own spending ceiling in code
registerExactEvmScheme(client, { signer: privateKeyToAccount(KEY) });
const payFetch = wrapFetchWithPayment(fetch, client);

const res = await payFetch("https://agent402.tools/api/llm", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    "model": "gpt-4o-mini",
    "messages": [
      {
        "role": "user",
        "content": "Say hello in one sentence."
      }
    ],
    "max_tokens": 64
  }),
});

Related tools