Make

Call TokenMix from any Make scenario with the HTTP Make a request module

What it is

Make, formerly Integromat, is a visual no-code automation platform from Celonis. You build scenarios by connecting modules on a canvas to move data between apps, call APIs, and add AI steps. It connects to thousands of services.

Before you begin

• A Make account (the free plan is enough to start).
• A TokenMix account and an API key from tokenmix.ai/dashboard/keys (sk-tm-...).
• The endpoint https://api.tokenmix.ai/v1/chat/completions.

Make's built-in OpenAI module is tied to OpenAI's own endpoint and has no custom base URL field. To use TokenMix, use the generic HTTP module instead, which can call any API.

Connect TokenMix

1. In your scenario, add a module and search for HTTP, then choose Make a request.
2. Set URL to https://api.tokenmix.ai/v1/chat/completions and Method to POST.
3. Add a header: name Authorization, value Bearer sk-tm-... (your TokenMix key). Set the body type to Raw with content type JSON (application/json).
4. In the request content, send a standard chat body, for example: {"model":"claude-sonnet-4.6","messages":[{"role":"user","content":"Hello"}]}
5. Run the module. The reply comes back in the response body; map choices[0].message.content into your next module.

Verify it works

Run the scenario once. A normal JSON response with a choices array means the call reached TokenMix. Check the spend in your TokenMix dashboard under usage.

FAQ

Why not the OpenAI module? Make's OpenAI module is bound to OpenAI's account and endpoint, with no place to enter a custom base URL. The HTTP module is the supported way to reach an OpenAI-compatible provider like TokenMix.

Which model should I use? For high-volume automations, a fast, low-cost model like deepseek-v4-flash or gemini-3-flash-preview works well; for harder reasoning, use claude-sonnet-4.6 or gpt-5.5.