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

# Clients and integrations

> Choose your client and follow the APIONE guide to configure keys, endpoints, and models.

APIONE provides OpenAI-, Anthropic-, and Gemini-compatible APIs. Pick the client you use, then follow its guide to fill in the address and model. The same field can mean a root URL, a Base URL, or a complete request URL depending on the software.

![APIONE client setup page](https://mintcdn.com/apione/pAFvFGkbmcRZirTZ/images/ui/api-keys-overview.png?fit=max\&auto=format\&n=pAFvFGkbmcRZirTZ\&q=85\&s=286a882ab53c81ad3bd16d56c6487dc9)

The APIONE API Key page shows the Base URL, protocol notes, model pricing, and client downloads.

## Before you start

1. Create an API Key in the [APIONE console](https://apione.ai/en/app/api-keys).
2. Choose a model in the [model directory](https://apione.ai/en/models) and confirm that your key can access it.
3. Copy the actual model ID. YOUR\_APIONE\_KEY and YOUR\_APIONE\_MODEL\_ID in the guides are placeholders; replace them before use.
4. Test a simple text conversation first, then test streaming, images, and tool calls one at a time.

## Coding assistants and command-line tools

> [card] [CC-Switch](https://apione.ai/docs/en/guides/cc-switch.md)
>
> Save APIONE configurations in a GUI and switch providers per tool.

> [card] [Claude Code](https://apione.ai/docs/en/guides/claude-code.md)
>
> Set an Anthropic-compatible address and use APIONE in the terminal and VS Code.

> [card] [Claude Desktop](https://apione.ai/docs/en/guides/claude-desktop.md)
>
> Configure the APIONE Anthropic Messages-compatible API.

> [card] [Cline](https://apione.ai/docs/en/guides/cline.md)
>
> Configure an OpenAI Compatible model provider in your editor.

> [card] [Codex CLI and desktop](https://apione.ai/docs/en/guides/codex.md)
>
> Configure APIONE through a custom Responses provider.

> [card] [Cursor](https://apione.ai/docs/en/guides/cursor.md)
>
> Set a custom OpenAI address and verify your chat model.

> [card] [DeepSeek Harness](https://apione.ai/docs/en/guides/deepseek-harness.md)
>
> Configure DSH with an API Key and a custom base URL.

> [card] [Gemini CLI](https://apione.ai/docs/en/guides/gemini-cli.md)
>
> Call APIONE models from the command line through the Gemini-compatible API.

> [card] [GitHub Copilot](https://apione.ai/docs/en/guides/github-copilot.md)
>
> Configure APIONE through custom model endpoints in VS Code.

> [card] [Hermes Agent](https://apione.ai/docs/en/guides/hermes-agent.md)
>
> Configure Hermes Agent with a custom OpenAI-compatible endpoint.

> [card] [Neovate Code](https://apione.ai/docs/en/guides/neovate-code.md)
>
> Add a custom provider to use APIONE in Neovate Code.

> [card] [OpenCode](https://apione.ai/docs/en/guides/opencode.md)
>
> Connect to APIONE with an OpenAI-compatible provider.

> [card] [Pi Coding Agent](https://apione.ai/docs/en/guides/pi-coding-agent.md)
>
> Configure Pi with a custom provider and API Key.

## Chat, notes, and workflows

> [card] [Cherry Studio](https://apione.ai/docs/en/guides/cherry-studio.md)
>
> Add an OpenAI-compatible service and chat with APIONE models on desktop.

> [card] [Dify](https://apione.ai/docs/en/guides/dify.md)
>
> Add models to your workspace and call APIONE from chatbots and workflows.

> [card] [Obsidian Copilot](https://apione.ai/docs/en/guides/obsidian.md)
>
> Configure APIONE chat models in your notes workspace.

> [card] [Open WebUI](https://apione.ai/docs/en/guides/open-webui.md)
>
> Add APIONE models through the administrator connection settings.

> [card] [RikkaHub](https://apione.ai/docs/en/guides/rikkahub.md)
>
> Add APIONE as a provider in the Android chat client.

> [card] [Sider](https://apione.ai/docs/en/guides/sider.md)
>
> Use APIONE in Sider versions that support a custom API.

## Agent deployment and shared configuration

> [card] [OpenClaw](https://apione.ai/docs/en/guides/openclaw.md)
>
> Add APIONE as a custom OpenClaw model provider.

> [card] [Alibaba Cloud OpenClaw](https://apione.ai/docs/en/guides/openclaw-alibaba.md)
>
> Verify the model connection on the server first, then connect the gateway and messaging channels.

> [card] [Network environments](https://apione.ai/docs/en/guides/network-environments.md)
>
> Troubleshoot connections on local machines, proxies, containers, and remote services.

> [card] [OAuth PKCE](https://apione.ai/docs/en/guides/oauth-pkce.md)
>
> Learn how APIONE authentication works today and where OAuth integration stops.

## Endpoint reference

| API | Common Base URL | Example request path |
| - | - | - |
| OpenAI Chat Completions | `https://api.apione.ai/v1` | `/v1/chat/completions` |
| [OpenAI Responses](https://apione.ai/docs/en/api/responses.md) | `https://api.apione.ai/v1` | `/v1/responses` |
| Anthropic Messages | `https://api.apione.ai` | `/v1/messages` |
| [Gemini](https://apione.ai/docs/en/api/gemini.md) | `https://api.apione.ai/v1beta` | `/v1beta/models/{model}:generateContent` |

If the client appends the API version automatically, enter the root URL. For example, Gemini CLI's GOOGLE\_GEMINI\_BASE\_URL uses `https://api.apione.ai`. The API version should appear only once in the final request path. Follow each guide for the exact values.

## Generic OpenAI-compatible setup

1. Open the model provider settings, choose a custom OpenAI-compatible provider, and name it APIONE.
2. Enter your APIONE API Key and use `https://api.apione.ai/v1` as the Base URL.
3. If a complete endpoint is required, enter `https://api.apione.ai/v1/chat/completions`.
4. If Host and Path are separate fields, enter `https://api.apione.ai` and `/v1/chat/completions`.
5. Add the actual model ID, save, and start a new conversation to test.

## Loomy

1. Open **Settings → Models → Provider configuration** and click **Add provider**.
2. Set the provider ID to `apione`, the display name to **APIONE**, the SDK mode to **OpenAI Compatible**, and the Base URL to `https://api.apione.ai/v1`.
3. Click **Next**, enter your APIONE API Key, click **Fetch from API**, select a model, then save and apply.

![Loomy custom provider name, compatible protocol, and APIONE Base URL](https://mintcdn.com/apione/zM1w3ic8px2lIvc_/images/integrations/screenshots/loomy-apione-provider.png?fit=max\&auto=format\&n=zM1w3ic8px2lIvc_\&q=85\&s=4109974fe40ce45d3d04432cc8c14f5e)

Example of adding APIONE as a custom provider in Loomy.

![Loomy masked API Key and Fetch from API button](https://mintcdn.com/apione/zM1w3ic8px2lIvc_/images/integrations/screenshots/loomy-api-key-models.png?fit=max\&auto=format\&n=zM1w3ic8px2lIvc_\&q=85\&s=eb337d082aba6464259741c3f57ee517)

The Loomy API Key field and the fetch models button.

## Zed

1. Open **AI settings → LLM Providers** and click **Add Provider**.
2. Add an OpenAI-compatible provider for Zed's built-in Agent. Set the provider name to **APIONE**, the API URL to `https://api.apione.ai/v1`, and the API Key to your APIONE key.
3. Add an available model ID. After saving, select the model in Zed Agent and send a test message.

![Add provider entry in Zed AI settings](https://mintcdn.com/apione/zM1w3ic8px2lIvc_/images/integrations/screenshots/zed-add-provider-entry.png?fit=max\&auto=format\&n=zM1w3ic8px2lIvc_\&q=85\&s=9949a31fb45a6415502799df87bcae71)

The LLM Providers section and Add Provider button in Zed.

## Kilo Code

1. Open the extension's **Settings → Providers → Custom provider**.
2. Set the Provider ID to `apione` and the Display name to **APIONE**. For Chat Completions models, set Provider API to **OpenAI Compatible**, the Base URL to `https://api.apione.ai/v1`, and the API Key to your APIONE key.
3. Fetch or manually add an available model, save, then select it in chat and send a test message.

![APIONE Base URL and API Key fields in the Kilo Code custom provider form](https://mintcdn.com/apione/zM1w3ic8px2lIvc_/images/integrations/screenshots/kilo-code-apione-provider.png?fit=max\&auto=format\&n=zM1w3ic8px2lIvc_\&q=85\&s=66ca33c68d7deed313b333fed5a1ce24)

Example of a custom provider configuration in Kilo Code.

## LobeHub

1. In **Settings → AI providers**, create a custom provider. Set the identifier to `apione` and the provider name to **APIONE**.
2. Set the request format to **OpenAI** and the proxy URL to `https://api.apione.ai/v1`, enter your APIONE API Key, and click **Create**.
3. Add an available model ID, select it in chat, and send a test message.

![OpenAI request format, APIONE proxy URL, and key fields in the LobeHub custom provider form](https://mintcdn.com/apione/l_tpyU_NU9GJriKq/images/integrations/screenshots/lobehub-apione-provider.png?fit=max\&auto=format\&n=l_tpyU_NU9GJriKq\&q=85\&s=8ae8ea8deb2b2f784bedb9499e986a15)

Example of a custom provider configuration in LobeHub. The provider identifier can be apione.

## ExtraBrain

1. Open **Settings → AI Providers → Custom Provider**.
2. Enter your APIONE API Key and the Base URL `https://api.apione.ai/v1`. Set **Model Name** to an actual model ID from the APIONE model directory.
3. Click **Validate** to check the configuration. Once it passes, set it as the primary or backup provider if needed.

![APIONE Base URL, model name, and Validate button in ExtraBrain](https://mintcdn.com/apione/l_tpyU_NU9GJriKq/images/integrations/screenshots/extrabrain-apione-provider.png?fit=max\&auto=format\&n=l_tpyU_NU9GJriKq\&q=85\&s=183e33ea1f2462dae435dc0d70da79fc)

Example of a custom provider configuration in ExtraBrain. Replace the placeholder model name with an actual model ID.

## Verify and troubleshoot

A "Test connection" button does not always use the same API as a real conversation. Send a real request and check it in your APIONE request logs. If something fails, check the final URL, protocol, model ID, key permissions, and balance in that order. Client subscription features, OAuth sign-in, and proprietary extensions are outside the scope of generic API compatibility.

Menu names may change between client versions. Find the equivalent fields in your current interface and use the APIONE values in this guide.

Related: [SDKs](https://apione.ai/docs/en/guides/sdks.md) · [Migrate an existing app](https://apione.ai/docs/en/guides/migration.md) · [Errors](https://apione.ai/docs/en/support/errors.md) · [FAQ](https://apione.ai/docs/en/support/faq.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.