Docs › Quickstart

Quickstart

Key, credit, first call: about five minutes from start to finish.

1. Get an API key

   Open Get API key and create a key. No email or card needed. Your key starts with ck_ and is shown only once, so save it in a password manager straight away. We only keep a fingerprint of it and can’t show it again.

   Store it in an environment variable so it stays out of your code:

   Shell

   export TOKENCENTER_API_KEY="ck_your_key_here"

2. Add credit

   New keys start with a $0 balance, and model calls need credit. On Pricing, pick a pack while you’re signed in with your key: Mini ($5), Maker ($19) or Studio ($59). Checkout runs on Stripe’s hosted page, and the credit is added to your balance once Stripe confirms the payment.

3. Make your first call

   Use any OpenAI-compatible client. The example uses Ministral 3B, one of the cheapest models in the catalog at $0.10 per million input and output tokens, so this call costs a tiny fraction of a cent.

   curlPythonJavaScript

   curl

   curl https://api.tokenctr.com/v1/chat/completions \
     -H "Authorization: Bearer $TOKENCENTER_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{
       "model": "mistralai/ministral-3b-2512",
       "messages": [{"role": "user", "content": "Say hello in five words."}],
       "max_tokens": 100
     }'

   Python

   # pip install openai
   import os
   from openai import OpenAI
   
   client = OpenAI(
       base_url="https://api.tokenctr.com/v1",
       api_key=os.environ["TOKENCENTER_API_KEY"],
   )
   
   resp = client.chat.completions.create(
       model="mistralai/ministral-3b-2512",
       messages=[{"role": "user", "content": "Say hello in five words."}],
       max_tokens=100,
   )
   print(resp.choices[0].message.content)
   print(resp.usage)

   JavaScript

   // npm install openai   (Node 18+, ES modules)
   import OpenAI from "openai";
   
   const client = new OpenAI({
     baseURL: "https://api.tokenctr.com/v1",
     apiKey: process.env.TOKENCENTER_API_KEY,
   });
   
   const resp = await client.chat.completions.create({
     model: "mistralai/ministral-3b-2512",
     messages: [{ role: "user", content: "Say hello in five words." }],
     max_tokens: 100,
   });
   console.log(resp.choices[0].message.content);
   console.log(resp.usage);

   Streaming isn’t available yet

   Send requests without stream, or with "stream": false. With "stream": true the API answers with an error event instead of a reply. If your tool streams by default, switch streaming off in its settings.

4. Check your balance

   Every successful call returns your remaining balance in the x-cosmos-credit-balance response header. You can also ask for it directly:

   curl

   curl https://api.tokenctr.com/v1/usage/today \
     -H "Authorization: Bearer $TOKENCENTER_API_KEY"

   credits_face_usd is your balance in US dollars, and rollup lists today’s calls, tokens and cost by model. The Account page shows the same numbers.

Next steps

• Browse the model catalog and copy any model ID, such as anthropic/claude-haiku-4.5 or mistralai/ministral-3b-2512.
• Read the API reference for request fields, response headers and account endpoints.
• Set max_tokens on every request. It keeps the temporary hold on your balance small; see how holds work.
• Coming from OpenRouter? See Migrate from OpenRouter.