> ## 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.

# LangBot

> Connect LangBot bot workflows to TokenLab models

## Overview

LangBot can call TokenLab through an OpenAI-compatible LLM requester.

<Note>
  **Type**: Bot Platform

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

  **Support Confidence**: Supported with scope limits
</Note>

## TokenLab Requester

If your LangBot build includes a **TokenLab** requester:

1. Open LangBot's model provider settings.
2. Add a provider with requester **TokenLab**.
3. Set the API key to your TokenLab key.
4. Select or create a model such as `gpt-5.4-mini`.
5. Attach the model to the pipeline or bot that should use it.

## OpenAI-Compatible Fallback

If TokenLab is not listed yet, use the OpenAI-compatible requester:

| Field     | Value                        |
| --------- | ---------------------------- |
| Requester | OpenAI Chat Completions      |
| Base URL  | `https://api.tokenlab.sh/v1` |
| API Key   | `sk-your-tokenlab-key`       |
| Model     | `gpt-5.4-mini`               |

## Recommended Models

* `gpt-5.4-mini` for most bot interactions
* `claude-sonnet-5` for longer writing and reasoning
* `deepseek-v4-flash` for fast coding and reasoning loops
* `gemini-3.5-flash` for fast general workflows