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

# Configure Third-party Apps

> Use APIXO in apps that support custom Claude, OpenAI, or Gemini endpoints

Many coding tools, agent frameworks, chat apps, and workflow builders let you configure a custom provider endpoint. If the app supports Claude, OpenAI, or Gemini compatible APIs, APIXO usually works by changing the base URL, API key, and model id.

This page intentionally uses a generic setup pattern. Tool-specific guides can be added later after each tool's current settings are verified.

## Quick checklist

| Field in the app     | What to choose                                                                   |
| -------------------- | -------------------------------------------------------------------------------- |
| Provider             | Choose the provider format the app already supports: Claude, OpenAI, or Gemini   |
| Base URL             | Use the matching APIXO base URL below                                            |
| API key              | Use your APIXO API key from the [dashboard](https://apixo.ai/dashboard/api-keys) |
| Environment variable | Prefer `APIXO_API_KEY` when the app supports env vars                            |
| Model                | Use a model id from [Supported Models](/docs/llm/supported-models)                    |
| Streaming            | Enable only if the app supports that provider's streaming format                 |

## Claude-compatible settings

Use this when the app asks for an Anthropic, Claude, or Claude-compatible endpoint.

| Setting                 | Value                           |
| ----------------------- | ------------------------------- |
| Base URL                | `https://llm.apixo.ai`          |
| API key                 | Your APIXO API key              |
| API key header          | `x-api-key`                     |
| Required version header | `anthropic-version: 2023-06-01` |
| Example model           | `claude-sonnet-4-6`             |

## OpenAI-compatible settings

Use this when the app asks for an OpenAI-compatible endpoint or custom OpenAI base URL.

| Setting        | Value                                  |
| -------------- | -------------------------------------- |
| Base URL       | `https://llm.apixo.ai/v1`              |
| API key        | Your APIXO API key                     |
| API key header | `Authorization: Bearer $APIXO_API_KEY` |
| Endpoint       | `/responses`                           |
| Example model  | `gpt-5.4`                              |

## Gemini-compatible settings

Use this when the app asks for a Gemini-compatible API endpoint.

| Setting        | Value                                    |
| -------------- | ---------------------------------------- |
| Base URL       | `https://llm.apixo.ai`                   |
| API key        | Your APIXO API key                       |
| API key header | `x-goog-api-key`                         |
| API path style | `/v1beta/models/{model}:generateContent` |
| Example model  | `gemini-3.1-pro-preview`                 |

## Common issues

* If the app validates model names, copy one from [Supported Models](/docs/llm/supported-models).
* If streaming fails, disable streaming and confirm a non-streaming request first.
* If Claude auth fails, confirm the app sends the `anthropic-version` header.
* If an OpenAI-compatible app does not support the Responses API, confirm which endpoint it calls before using APIXO.

<Info>
  APIXO does not require a special integration for each app. The app only needs to send provider-compatible requests to a custom base URL.
</Info>