Configuration

The first step to using Auto-Coder.Chat is configuring the model you want to use. Navigate to your project directory and launch the CLI:

cd your-project
auto-coder.chat

View Available Models

Run the following command to list all available models:

/models /list

You'll see the built-in model list with provider, description, pricing, and API Key status:

Tab Completion: Type /mo and press Tab to auto-complete to /models. Sub-commands also support Tab completion.

Configure Model API Key

Use the /models command to set an API Key for a model:

/models <model_name> <YOUR_API_KEY>

For example, to configure the Volcengine Coding Plan DeepSeek V3.2:

/models volcengine/coding/deepseek-v3.2  YOUR_API_KEY

The first argument is the model name (e.g., volcengine/coding/deepseek-v3.2), and the second is your API Key.

Coding Plan vs Regular API Keys

Models with coding in their name use the provider's Coding Plan — you'll need the corresponding Coding Plan API Key.

Quick rule: coding in the name → Coding Plan Key; no coding → standard official API Key.

Configure Vision Model

Auto-Coder.Chat supports a vision model (VL Model) for understanding images, screenshots, and other visual content. Use the /conf command to configure it:

/conf vl_model:volcengine/doubao-seed-1-6-thinking

The recommended model is volcengine/doubao-seed-1-6-thinking, which excels at image understanding and reasoning.

The vision model also requires a configured API Key. Make sure you've set up the API Key for the model via the /models command first.

Next Steps

Once your model is configured, see Basic Usage to learn about daily usage.