One API for
Top AI Models.

TheOldAPI provides a single API across a curated selection of top AI models. No pay-as-you-go, just a simple $7/mo subscription.

100
Models
40
Providers
OpenAI
Compatible
$7
/mo flat
OpenAI Anthropic Google Gemini DeepSeek Meta Llama Mistral xAI Grok Cohere Qwen Perplexity HuggingFace Together AI OpenAI Anthropic Google Gemini DeepSeek Meta Llama Mistral xAI Grok Cohere Qwen Perplexity HuggingFace Together AI

Explore and use powerful models

We provide API-based inference across leading text models — some with image-input (vision) support — through one OpenAI-compatible API.

Why choose TheOldAPI?

There are many reasons to choose us, but here are some of the most important ones.

OpenAI Compatible

Drop-in replacement. Works with any OpenAI SDK, LangChain, or HTTP client. Just change your base URL.

Fast Routing

Direct provider connections with connection pooling and minimal routing overhead.

Multi-Provider Failover

If one backend goes down, we automatically route your request to the next available provider.

Tool & Streaming Support

Supports common features such as streaming, tool/function calling, and image input (vision) where available from the selected model/provider.

Top Models

GPT-5.3, Claude Opus, Gemini Pro, Grok, DeepSeek, and many more — all from one API.

Anthropic Format

Native support for both OpenAI and Anthropic message formats via /v1/messages.

Live model explorer

Model Explorer

Browse supported model routes. Filter by provider or search by name. Model availability, routing, and behavior may vary — listed names are routing labels and may not guarantee access to a specific provider or version.

100 / 100 models

Loading models...

Simple pricing, powerful models

No pay-as-you-go. No hidden fees. Includes 50M adjusted tokens per day, resetting daily.

Monthly
$ 7 /month

Full access with a 50M adjusted-token daily limit

  • 100+ supported model routes
  • 50M adjusted tokens/day
  • 30 req/min rate limit
  • Streaming (when supported)
  • Multi-provider routing and fallback where available
  • Tool/function calling (when supported)
  • Vision (image input) where supported
  • Anthropic /v1/messages
Enterprise
Custom

For teams with high-volume needs

  • Everything in Pro
  • Custom rate limits
  • Dedicated support
  • Multi-provider routing and fallback where available
  • Usage analytics
  • Invoice billing
No Hidden Fees
Cancel Anytime

Why TheOldAPI?

Direct API Access
$50-200+/1M tokens
  • Pay per token
  • Multiple API keys needed
  • No failover
  • Manage each provider separately
TheOldAPI
$7/mo
  • 50M adjusted tokens/day
  • One API key for all
  • Automatic failover
  • 100+ models, one dashboard

Frequently Asked Questions

What are adjusted tokens?

You get 50M adjusted tokens per day, resetting at midnight UTC. Each real token counts against this budget by a model multiplier — cheaper models count close to 1x, while premium models count for more, so they draw down your daily budget faster. Every model's current multiplier is shown as a cost badge on the Models page. Only tokens you actually use (input + output) are counted, and you can mix models freely.

How much can I use for $7/month?

The 50M adjusted-token allowance resets daily, so there's no pay-as-you-go and no overage charges. If you hit the daily limit, requests pause until the next midnight-UTC reset. Your live usage is always shown on your dashboard.

Is this OpenAI-compatible?

Yes! Point your OpenAI SDK base URL to our API and it works instantly.

Can I cancel anytime?

Absolutely. Cancel anytime and retain access until the end of your billing period.

How do I get my API key?

Subscribe to a paid plan to unlock your API key. Checkout is processed securely by Polar, and your key appears in your dashboard automatically once your subscription is active.

Welcome Back

Sign in to access your dashboard

Don't have an account?

Dashboard

Welcome back, user

All Systems Operational
Models 100+
Providers 40+
Plan $7/mo

Membership

Free
Your Plan
Free
$0/mo
  • No API access
  • No models
  • No streaming
Upgrade to
Pro
$7/mo
  • 50M adjusted tokens/day
  • 100+ AI models
  • Streaming when supported

Head to the pricing page to subscribe. Checkout is processed securely by Polar and your key appears here automatically once active.

Your API Key

Subscribe to a paid plan to unlock your API key.

Quick Start

from openai import OpenAI

client = OpenAI(
    base_url="https://your-api.com/v1",
    api_key="YOUR_API_KEY"
)

resp = client.chat.completions.create(
    model="gpt-5.4",
    messages=[{"role": "user",
               "content": "Hello!"}]
)
print(resp.choices[0].message.content)

Endpoints

POST /v1/chat/completions
POST /v1/messages
GET /v1/models
POST /v1/messages/count_tokens
Developer docs

Documentation

Everything you need to integrate TheOldAPI into your application.

Quick Start

TheOldAPI is fully compatible with the OpenAI SDK. Just change your base_url and you're ready to go.

1. Install the SDK

pip install openai

2. Make your first request

from openai import OpenAI

client = OpenAI(
    base_url="/v1",
    api_key="your-api-key"
)

response = client.chat.completions.create(
    model="gpt-5.4",
    messages=[
        {"role": "user", "content": "Hello!"}
    ],
    stream=True
)

for chunk in response:
    print(chunk.choices[0].delta.content, end="")

Authentication

All API requests require an API key passed via the Authorization header:

Authorization: Bearer YOUR_API_KEY

Get your API key from the Dashboard after signing up.

API Endpoints

POST /v1/chat/completions

Create a chat completion. Supports all OpenAI-compatible parameters including stream, tools, temperature, max_tokens, etc.

GET /v1/models

List all available models. Returns model IDs and provider information.

POST /v1/messages

Anthropic-compatible messages endpoint. Supports Claude-format requests natively.

POST /v1/messages/count_tokens

Count tokens for an Anthropic-format messages request before sending.

Rate Limits

Plan Requests/min Models
Pro ($7/mo) 30 100+ routes
Enterprise Custom 100+ routes

Rate limit headers are included in every response: X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset.

Streaming

Set "stream": true in your request body to receive Server-Sent Events (SSE). Each event contains a chunk of the response.

curl /v1/chat/completions \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-5.4","stream":true,
       "messages":[{"role":"user","content":"Hi"}]}'

Node.js Example

import OpenAI from "openai";

const client = new OpenAI({
    baseURL: "/v1",
    apiKey: "your-api-key",
});

const completion = await client.chat.completions.create({
    model: "claude-sonnet-4-20250514",
    messages: [{ role: "user", content: "Hello!" }],
});

console.log(completion.choices[0].message.content);

Join our Discord community

Join our Discord community to connect with other developers, get help, and stay updated on the latest TheOldAPI news.

Join Discord
theoldapi