Your first API request with CheapTokens

This guide takes you from a new account to a working request. You need an email address, a small balance and a terminal. The API is OpenAI-compatible, so if you have called OpenAI before, only two things change: the base URL and the key.

What you need#

• An account. Sign-up is by email — no password, we send you a sign-in link.
• A balance. There is no subscription: you top up a USD balance and every request is charged per token at the model's fixed rate.
• An API key from your dashboard.

Step 1. Create an account and top up#

Open the sign-up page, enter your email and follow the link from the letter. In the dashboard press Top up and choose how to pay:

• Crypto — from $5; a 5% deposit fee applies.
• Card — through a partner, who charges its own fee.

Step 2. Copy your API key#

As soon as your balance is above zero — after a top-up or a referral bonus — the dashboard creates your API key and shows it once. Copy it right away and keep it out of your code, for example in an environment variable:

export CHEAPTOKENS_API_KEY="paste-your-key-here"

If the key leaks, press Rotate: you get a new key and the old one stops working.

Step 3. Pick a model#

Requests need the full model ID — with the prefix before the slash, exactly as the API lists it. The catalog at the time of writing (October 2026):

The catalog can change, so the source of truth is the API itself. This request needs no key:

curl https://api.cheaptokens.dev/v1/models

Step 4. Send the request#

curl#

curl https://api.cheaptokens.dev/v1/chat/completions \
  -H "Authorization: Bearer $CHEAPTOKENS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-ai/DeepSeek-V4-Flash-0731",
    "messages": [{"role": "user", "content": "Say hello in three languages"}]
  }'

A successful answer comes back in the standard Chat Completions format. A shortened example — your text, IDs and numbers will differ, and the response may carry extra service fields:

{
  "id": "chatcmpl-…",
  "object": "chat.completion",
  "model": "deepseek-ai/DeepSeek-V4-Flash-0731",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "Hello! Hola! Bonjour!" },
      "finish_reason": "stop"
    }
  ],
  "usage": { "prompt_tokens": 14, "completion_tokens": 9, "total_tokens": 23 }
}

• The text is in choices[0].message.content.
• Token usage is in usage: prompt_tokens — input, completion_tokens — output. These two numbers are what the request is charged for.

Python#

Install the official SDK with pip install openai and point it at our base URL:

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.cheaptokens.dev/v1",
    api_key=os.environ["CHEAPTOKENS_API_KEY"],
)
resp = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V4-Flash-0731",
    messages=[{"role": "user", "content": "Say hello in three languages"}],
)
print(resp.choices[0].message.content)
print("tokens in/out:", resp.usage.prompt_tokens, resp.usage.completion_tokens)

JavaScript#

Install the SDK with npm install openai:

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.cheaptokens.dev/v1",
  apiKey: process.env.CHEAPTOKENS_API_KEY,
});
const resp = await client.chat.completions.create({
  model: "deepseek-ai/DeepSeek-V4-Flash-0731",
  messages: [{ role: "user", content: "Say hello in three languages" }],
});
console.log(resp.choices[0].message.content);

Anthropic-style clients#

Clients built for the Anthropic Messages API use the root address without /v1 — https://api.cheaptokens.dev — and send the key in the x-api-key header. The Anthropic SDKs add the path and the header themselves; you only set the base URL and the key.

If something goes wrong#

• 401 Invalid API key — the key is mistyped or was rotated. Check that you pasted it whole; the dashboard shows a key only once, so if it is lost, press Rotate to get a new one.
• 402 — the balance is used up, or the request went out without the Authorization header.
• 400 Model "…" not found — use the full model ID from /v1/models, including the prefix before the slash.
• 404 — check the address: OpenAI-style clients need /v1 at the end of the base URL.

More cases and fixes: Common connection errors and how to fix them.

What it costs#

Each request is charged for input and output tokens separately, at the model's rate per million tokens. Current prices — the same data as on the pricing page:

Next step: connect a coding agent, for example Cline, or learn to estimate what an AI agent will cost.

Recommending CheapTokens to other developers? The referral program credits you a percentage of their top-ups.