One endpoint · every provider

One API for all AI models

AivoKey unifies many providers behind one endpoint, with per-model budgets and transparent pricing per 1M tokens.

Change the base URL. Keep your code.

Point any [OI]- or Anthropic-compatible client at one gateway. The router handles provider selection, retries, failover, and metering — you just name the model.

Chat CompletionsPOST /v1/chat/completions
ResponsesPOST /v1/responses
MessagesPOST /v1/messages

quickstart.ts

import [OI] from "openai";

const client = new [OI]({
  baseURL: "https://api.aivokey.dev/v1",
  apiKey: process.env.AIVOKEY_API_KEY,
});

const res = await client.chat.completions.create({
  model: "deepseek-v4-pro",
  messages: [{ role: "user", content: "Hello" }],
});

5Models

5Providers

3Wire formats

5h / 7d / 1moBudget windows

Multi-provider routing

One endpoint for all models. The router picks the best upstream automatically, with cooldown and failover when an upstream is unhealthy.

Budget per model

Set a monthly allowance per model or a global pool. Exhausted? Remaining budget on other models can still be used according to your plan.

One API key

One key for all providers. Switch models by simply changing the model name in the request — without changing credentials.

Pricing that scales with tokens

Subscribe from $10/month for a per-model allowance, or pay as you go from a prepaid balance. No seats, no per-developer fees.

See plans Browse 5 models →

Per-model budgets5-hour, weekly, and monthly windows.

Automatic failoverUnhealthy upstreams are skipped.

Zero data retentionAvailable on most models by default.

No lock-inBring any other provider alongside us.

Start routing in a few minutes.

Create an account, mint a key, and point your client at the gateway.

Create an account Read the docs