Cheaper Inference

A Keak company

Docs Log in Create account
Drop-in SDK example Compatible
Only change these two lines 
const client = new OpenAI({
  apiKey: "ci_your_key",
  baseURL: "https://api.cheaperinference.com/v1"
});

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

Keep your existing OpenAI SDK and chat completion calls.

Cut AI API costs by 30% with one endpoint

Use OpenAI, Anthropic, and Google models through a drop-in OpenAI-compatible API. No contract, no migration required.

Change your endpoint. Keep your existing request format.

Start saving → View models & pricing
No credit card required OpenAI SDK compatible No commitment

How it works

Switch in minutes. Save on every request.

Keep your existing integration, route through one endpoint, and monitor usage from your wallet dashboard.

01

Replace your base URL

Keep your existing OpenAI SDK and request format.

02

Choose any supported model

Route requests to OpenAI, Anthropic, and Google models from one API.

03

Track spend and savings

Usage is deducted from your wallet at the discounted token rate.

Catalog

Supported models and pricing

Access

Ready to start saving?

Create an account, fund your wallet, and issue your first API key.

Account

Sign in or create an account

Docs

Documentation

Everything you need to create an account, choose a model, send requests, and understand billing.

Quick start

  1. Create an account and verify your email.
  2. Create an API key from the dashboard and copy it once.
  3. Set your OpenAI SDK baseURL to the endpoint below.
  4. Put a supported model name in the model field.
  5. Send your first chat completion request and review usage in History.

Base URL

https://api.cheaperinference.com/v1

Authentication

Authorization: Bearer ir_live_YOUR_API_KEY

Example: cURL chat completion

curl https://api.cheaperinference.com/v1/chat/completions \
  -H "Authorization: Bearer ir_live_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-4.6",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Example: JavaScript SDK

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "ir_live_YOUR_API_KEY",
  baseURL: "https://api.cheaperinference.com/v1"
});

const response = await client.chat.completions.create({
  model: "gpt-5.4",
  messages: [{ role: "user", content: "Hello! Who are you?" }]
});

Example: Python SDK

from openai import OpenAI

client = OpenAI(
    api_key="ir_live_YOUR_API_KEY",
    base_url="https://api.cheaperinference.com/v1",
)

response = client.chat.completions.create(
    model="claude-opus-4.6",
    messages=[{"role": "user", "content": "Hello! Who are you?"}],
)

List models

curl https://api.cheaperinference.com/v1/models \
  -H "Authorization: Bearer ir_live_YOUR_API_KEY"

Streaming

Streaming responses are not supported yet. Do not send "stream": true; leave it out or set "stream": false. Requests that ask for streaming return a clear 400 error.

Available model names and pricing

Copy an exact model id into the model field. Prices are per 1M tokens.

Billing and wallet

  • Wallet balance is checked before a request starts.
  • Input and output tokens are deducted after each completed request.
  • Add funds manually or enable auto-recharge from the dashboard.

Common errors

  • 401 Invalid or missing API key.
  • 402 Insufficient wallet balance.
  • 400 Unsupported model or invalid request body.
  • 502 Provider request failed; retry or contact support if it continues.

FAQ

  • Do I need a contract? No. Create an account, add funds, and start sending requests.
  • Do I need to migrate code? Usually only the API key and base URL change.
  • Where do I pick the model? Pass the model id in every request.
  • Can I track savings? Yes. History shows request count, tokens, cost, and estimated savings.

Successful response

{
  "output": [{
    "role": "assistant",
    "content": [{"type": "output_text", "text": "Hello!"}]
  }],
  "usage": {
    "input_tokens": 12,
    "output_tokens": 3,
    "total_tokens": 15
  }
}

Security and keys

  • Store API keys in environment variables, not frontend code.
  • Create separate keys for production, staging, and local testing.
  • If a key is exposed, delete it and create a new one.
  • Use History to audit request volume, token usage, and spend.

Support

Use the Contact tab for billing, API, or account questions. We typically answer in under 12h. For platform availability, check the Status link in the footer.