TypingMind

An advanced bring-your-own-key chat UI that works with any OpenAI-compatible model.

What TypingMind is

TypingMind is an advanced, bring-your-own-key chat interface for AI models. It runs in your browser and stores your key and chat history locally. By adding a custom model that points at TokenMix, you can chat with Claude, GPT, DeepSeek and more on your own key. It suits people who want one polished chat interface across many models.

Before you start

• Open TypingMind at typingmind.com. It runs in the browser, and you can also install it as an app.
• Sign up at tokenmix.ai, open the Console, go to API Keys, and create a key that starts with sk-tm-.

Connect TokenMix

Two details here are easy to miss: the Endpoint must be the full URL ending in /chat/completions, and the key goes in a custom header rather than a separate key field.

1. Go to Settings, open Models, and click Add Custom Model.
2. Name: any label, for example TokenMix.
3. Endpoint: enter https://api.tokenmix.ai/v1/chat/completions, the full path, not just /v1.
4. Model ID: enter a TokenMix model id, for example claude-sonnet-4.6.
5. Expand Add Custom Headers. In the first box enter Authorization. In the second box enter the word Bearer, then a space, then your sk-tm- key.
6. Click Test. When it reports that the endpoint is working, click Add Model.

Check it works

Start a new chat, pick your TokenMix model from the model list, and send a short message such as "hello". A reply means it is connected. You can see the spend in the TokenMix Console.

Common questions

Test fails or no reply: the Endpoint must be the full https://api.tokenmix.ai/v1/chat/completions, and the key must be in a header named Authorization with the value Bearer followed by your key. A plain /v1 endpoint will not work here.

Add more models: repeat Add Custom Model for each model id you want, all using the same endpoint and header.

Which models to pick: any text model works. Claude Sonnet 4.6, GPT-5.4 and DeepSeek V4 Pro are good starting points.