> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tokenlab.sh/llms.txt
> Use this file to discover all available pages before exploring further.

# DeepChat

> Connect DeepChat to TokenLab with an OpenAI-compatible base URL

## Overview

DeepChat works with TokenLab through a standard OpenAI-compatible provider configuration.

<Note>
  **Type**: Chat App

  **Primary Path**: OpenAI-compatible chat completions

  **Support Confidence**: Supported path
</Note>

## Configure DeepChat

1. Open **Settings** -> **Model Provider**
2. Add or edit an OpenAI-compatible provider
3. Enter:
   * **Provider name**: TokenLab
   * **Base URL**: `https://api.tokenlab.sh/v1`
   * **API Key**: `sk-your-tokenlab-key`
4. Add a model such as `gpt-5.4-mini`
5. Save and test a chat.

## Model Choices

Start with `gpt-5.4-mini` for a low-friction default. Use `claude-sonnet-5`, `gpt-5.4`, or `deepseek-v4-pro` when the conversation needs stronger reasoning.

## Troubleshooting

* Use the base URL, not the full `/chat/completions` path, unless DeepChat explicitly asks for a full endpoint URL.
* Remove trailing spaces from the API key.
* If a model is not listed, add the exact TokenLab model id manually.