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

# Using in Codex CLI

> Configure APIAny.AI API keys, base URLs, and recommended models in Codex CLI.

[Codex CLI](https://github.com/openai/codex) can connect to APIAny.AI through environment variables or local configuration files. It is useful for terminal, editor, and automated development workflows.

## Prerequisites

* Codex CLI is installed.
* You have created an API key in the [APIAny.AI dashboard](https://apiany.ai/dashboard/api-keys).
* You have selected a model ID.

## Install

```bash theme={null}
npm install -g @openai/codex
```

## Configure the API key

Add this to `~/.codex/auth.json`:

```json theme={null}
{
  "OPENAI_API_KEY": "sk-xxxxxxxxxxxx"
}
```

## Configure config.toml

Add this to `~/.codex/config.toml`:

```toml theme={null}
model = "gpt-5.5"
model_provider = "apiany"

[model_providers.apiany]
name = "APIAny.AI"
base_url = "https://apiany.ai/v1"
wire_api = "responses"
requires_openai_auth = true
```

Restart your terminal, then run `codex` or test with:

```bash theme={null}
codex "Introduce yourself in one sentence"
```

## Recommended models

| Model ID            | Best for                                                                        |
| ------------------- | ------------------------------------------------------------------------------- |
| `gpt-5.5`           | Best first choice for complex reasoning, coding, and production workflows.      |
| `gpt-5.4`           | Strong general-purpose model for advanced chat and agent tasks.                 |
| `gpt-5.4-mini`      | Balanced speed, quality, and cost for everyday usage.                           |
| `gpt-4o-mini`       | Fast and economical choice for simple chat, translation, and lightweight tasks. |
| `claude-sonnet-4-6` | Claude-compatible model ID for long-form writing and coding workflows.          |
| `gemini-3.5-flash`  | Gemini-compatible model ID for fast multimodal usage.                           |

## Troubleshooting

### 401 or authentication errors

Make sure the API key is copied completely, has no extra spaces, and is enabled in the APIAny.AI dashboard.

### The model does not appear in the app

Enter the model ID manually and confirm it is available in the [model catalog](/models/chat-models). Some tools do not fetch model lists automatically.

### Which Base URL should I use?

OpenAI-compatible settings usually use `https://apiany.ai/v1`; Claude-compatible settings usually use `https://apiany.ai`; tools that ask for a full endpoint use `https://apiany.ai/v1/chat/completions`.