How to connect Claude Code to the Tokenator API

What you end up with

Once configured, Claude Code stops talking to its default provider and sends every request to Tokenator. Billing comes out of your key's token limit, and the list of available models is defined by the key rather than by the tool.

Requirements and supported operating systems

• Windows 10/11, macOS 12+, or a modern 64-bit Linux.
• Outbound HTTPS access to the Tokenator domain.
• A Tokenator API key that looks like sk-your-tokenator-key.

Installing via npm requires Node.js LTS. Codex CLI also ships standalone binaries, in which case Node is not required.

Installing Claude Code

npm install -g @anthropic-ai/claude-code

Verify the binary is on your PATH: claude --version.

Setting up Tokenator

1. Register in your account and buy a token bundle.
2. The key appears in the account right after the payment is confirmed — copy it whole, including the prefix.
3. The examples below use the placeholder sk-your-tokenator-key. Never publish a real key in repositories or screenshots.

Configuration through settings.json

Claude Code reads its settings from settings.json: the Tokenator address and the key are written there once, and the tool picks them up on every launch — nothing has to be typed into the terminal.

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.tokenator.top/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "sk-your-tokenator-key",
    "ANTHROPIC_MODEL": "claude-opus-4-6"
  }
}

If the file does not exist yet, create it together with the .claude folder in your home directory. A ready-made settings.json with your key and model list can be downloaded from the key page, which opens from your account. If you have already signed in to Claude Code with an Anthropic subscription, the key from the file takes precedence; remove the env block to go back to the subscription.

Your first request

Before starting the tool it helps to confirm that the key and the model both work. A single curl request answers both questions.

curl https://api.tokenator.top/anthropic/v1/messages \
  -H "x-api-key: sk-your-tokenator-key" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-4-6",
    "max_tokens": 64,
    "messages": [{"role": "user", "content": "ping"}]
  }'

cd /path/to/project
claude

A successful response means the key is valid, the model is allowed for it and the upstream is reachable. Usage appears in your account right after the request.

Switching models

The model comes from ANTHROPIC_MODEL. Set a new value and restart the agent to switch.

Examples of model IDs available right now:

• claude-haiku-4-5
• claude-opus-4-6
• claude-opus-4-7
• claude-sonnet-4-6
• gpt-5.5
• deepseek-v4-pro

Current IDs and specs live in the AI model catalog, which also shows which models are available right now. The model comparison helps choose between close options.

Configuring reasoning

The key's reasoning level is switched in your account and applied on the Tokenator side — Claude Code needs no separate setting for it. With reasoning disabled for the key, requests are handled without extended thinking.

Note that thinking is billed as output tokens: the higher the level, the faster the key limit is consumed.

Common errors and how to fix them

FAQ

Read next