# Cherry Studio Integration with OfoxAI Configuration Guide

Source: https://ofox.ai/docs/integrations/cherry-studio

[Cherry Studio](https://cherry-ai.com) is a cross-platform AI desktop client (supports Windows, macOS, Linux) with multi-model conversations, knowledge base, and workflow support — one of the most feature-rich open-source AI clients available.

By configuring OfoxAI as your provider, you can use GPT, Claude, Gemini, DeepSeek, and 100+ other models in Cherry Studio with just one API Key.

## Protocol Overview

Cherry Studio supports connecting to OfoxAI via four protocols:

**OpenAI Chat** — Based on `/v1/chat/completions`, for GPT series models such as `openai/gpt-6.1-sol`, `openai/gpt-6-luna`.

**OpenAI Response** — Based on `/v1/responses`, for GPT series models such as `openai/gpt-6.1-sol`, `openai/gpt-6-luna`.

**Claude** — Based on `/anthropic/v1/messages`, for Claude series models such as `anthropic/claude-sonnet-5.5`, `anthropic/claude-opus-5.5`.

**Gemini** — Based on `/gemini/v1beta/models`, for Gemini series models such as `google/gemini-3.8-flash`, `google/gemini-3.5-flash-lite`.

## Prerequisites

- An OfoxAI account with an API Key ([Get one here](https://app.ofox.ai))
- Cherry Studio installed ([Download](https://cherry-ai.com))

## Configuration Steps

### Step 1: Open Settings

Launch Cherry Studio and click the **Settings** icon in the top-right corner.

![Open Cherry Studio Settings](https://ofox.ai/screenshots/cherry-studio/01-open-settings.png)

### Step 2: Add a Model Provider

Go to **Model Provider** in the left sidebar, scroll to the bottom, and click **+ Add**. In the dialog, enter a provider name and select the corresponding provider type.

**OpenAI Chat**

Select `OpenAI` as the provider type and click OK.

![Add OpenAI Chat Provider](https://ofox.ai/screenshots/cherry-studio/02-add-chat.png)

**OpenAI Response**

Select `OpenAI` as the provider type (enable Response mode on the config page later) and click OK.

![Add OpenAI Response Provider](https://ofox.ai/screenshots/cherry-studio/02-add-response.png)

**Claude**

Select `Anthropic` as the provider type and click OK.

![Add Claude Provider](https://ofox.ai/screenshots/cherry-studio/02-add-claude.png)

**Gemini**

Select `Gemini` as the provider type and click OK.

![Add Gemini Provider](https://ofox.ai/screenshots/cherry-studio/02-add-gemini.png)

### Step 3: Enter API Configuration

Fill in the **API Key** and **API Host** on the provider's configuration page.

**OpenAI Chat**

| Field | Value |
|-------|-------|
| **API Key** | Your OfoxAI API Key |
| **API Host** | `https://api.ofox.ai/v1` |

![OpenAI Chat Configuration](https://ofox.ai/screenshots/cherry-studio/03-config-chat.png)

**OpenAI Response**

| Field | Value |
|-------|-------|
| **API Key** | Your OfoxAI API Key |
| **API Host** | `https://api.ofox.ai/v1` |

![OpenAI Response Configuration](https://ofox.ai/screenshots/cherry-studio/03-config-response.png)

**Claude**

| Field | Value |
|-------|-------|
| **API Key** | Your OfoxAI API Key |
| **API Host** | `https://api.ofox.ai/anthropic` |

![Claude Configuration](https://ofox.ai/screenshots/cherry-studio/03-config-claude.png)

**Gemini**

| Field | Value |
|-------|-------|
| **API Key** | Your OfoxAI API Key |
| **API Host** | `https://api.ofox.ai/gemini` |

![Gemini Configuration](https://ofox.ai/screenshots/cherry-studio/03-config-gemini.png)

Make sure the toggle in the top-right corner of the page is **ON**, otherwise the provider will not appear in the model selector.

### Step 4: Add Models

Click the **Manage** button in the model section to automatically fetch the model list, then click **+** next to a model to add it to the enabled list.

**OpenAI Chat**

![OpenAI Chat Model List](https://ofox.ai/screenshots/cherry-studio/04-model-chat.png)

**OpenAI Response**

![OpenAI Response Model List](https://ofox.ai/screenshots/cherry-studio/04-model-response.png)

**Claude**

![Claude Model List](https://ofox.ai/screenshots/cherry-studio/04-model-claude.png)

**Gemini**

![Gemini Model List](https://ofox.ai/screenshots/cherry-studio/04-model-gemini.png)

### Step 5: Verify Configuration

Click the **Check** button and select any model to test. A passing result means the configuration is successful.

**OpenAI Chat**

![OpenAI Chat Verification](https://ofox.ai/screenshots/cherry-studio/05-verify-chat.png)

**OpenAI Response**

![OpenAI Response Verification](https://ofox.ai/screenshots/cherry-studio/05-verify-response.png)

**Claude**

![Claude Verification](https://ofox.ai/screenshots/cherry-studio/05-verify-claude.png)

**Gemini**

![Gemini Verification](https://ofox.ai/screenshots/cherry-studio/05-verify-gemini.png)

## Start Using

Return to the main interface, click the model selector at the top of the chat, and choose a model under the corresponding provider group to start chatting.

![Select Model to Start Chat](https://ofox.ai/screenshots/cherry-studio/06-start-chat.png)

## Image Generation

Cherry Studio supports generating images via the Gemini protocol using OfoxAI's image generation models, directly within conversations.

### Step 1: Add a Gemini Provider

Go to **Model Provider** and click **+ Add** at the bottom.

![Add Provider](https://ofox.ai/screenshots/cherry-studio/img-01.jpg)

In the dialog, select **Gemini** as the provider type and click OK.

![Select Gemini Type](https://ofox.ai/screenshots/cherry-studio/img-02.jpg)

### Step 2: Configure and Fetch Image Models

Set the **API Host** to `https://api.ofox.ai/gemini`, then click **Fetch model list** and add the following image generation models:

- `Google: Nano Banana Pro (Gemini 3 Pro Image Preview)`
- `Google: Nano Banana (Gemini 2.5 Flash Image)`
- `Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)`

![Fetch Image Models](https://ofox.ai/screenshots/cherry-studio/img-03.jpg)

### Step 3: Edit Assistant and Configure Image Parameters

Right-click an assistant on the main interface and select **Edit Assistant**.

![Edit Assistant](https://ofox.ai/screenshots/cherry-studio/img-04.jpg)

On the edit page:
1. Set the **Default Model** to the image generation model you just added
2. Click **+ Add Parameter** and add the following two custom parameters:

| Parameter | Type | Value |
|-----------|------|-------|
| `responseModalities` | JSON | `["IMAGE"]` |
| `imageConfig` | JSON | `{"aspectRatio": "1:1", "imageSize": "1K"}` |

You can adjust the aspect ratio and resolution in `imageConfig`. For 2K images, change `imageSize` to `"2K"`.

![Configure Image Parameters](https://ofox.ai/screenshots/cherry-studio/img-05.jpg)

### Step 4: Start Generating

After saving the configuration, enter an image description in the chat to generate images.

![Image Generation Result](https://ofox.ai/screenshots/cherry-studio/img-06.jpg)

## New Cherry Studio: One Provider, Multiple Protocols

The latest Cherry Studio lets you set a separate request address for each protocol inside a single provider. Add one OfoxAI provider and you can use GPT, Claude, and Gemini models together, plus image generation and image editing.

### Step 1: Fill in the Request Configuration

After entering the API key on the provider page, open **Request configuration**, fill in each protocol address from the table below, and click **Save**:

| Protocol / Field | Address |
|------|------|
| **OpenAI (default)** | `https://api.ofox.ai/v1` |
| **Anthropic** | `https://api.ofox.ai/anthropic` |
| **Gemini** | `https://api.ofox.ai/gemini` |
| **OpenAI Responses** | `https://api.ofox.ai/v1` |
| **Image Generation Base URL** | `https://api.ofox.ai/v1` |
| **Image Edit Base URL** | `https://api.ofox.ai/v1` |

![Request configuration: OpenAI, Anthropic, Gemini, and OpenAI Responses addresses](https://ofox.ai/screenshots/cherry-studio/new-01-request-config.jpg)

![Request configuration: image generation and image edit Base URLs](https://ofox.ai/screenshots/cherry-studio/new-02-request-config-image.jpg)

**Set as default** decides which protocol regular chats use by default. **Image Generation Base URL** maps to `/images/generations` and **Image Edit Base URL** maps to `/images/edits`; leave them empty to fall back to the default chat endpoint's Base URL. The screenshots use the mirror `api.ofox.io`, which works from mainland China; use `api.ofox.ai` on other networks.

### Step 2: Chat with Multiple Models

After adding models, you can switch between Gemini, GPT, Claude, and other vendors' models under the same provider. The screenshot shows Gemini 3.8 Flash, GPT-6 Luna, and Claude Sonnet 5.5 in turn:

![Chatting with Gemini, GPT, and Claude models under one provider](https://ofox.ai/screenshots/cherry-studio/new-05-multi-protocol-chat.jpg)

### Step 3: Image Generation and Editing

After adding an image model (such as `openai/gpt-image-2`), select it on the drawing page and enter a prompt to generate an image:

![Generating an image with GPT-Image-2](https://ofox.ai/screenshots/cherry-studio/new-03-image-generate.jpg)

Upload a reference image and enter an edit instruction (for example, "turn it into a dog") to edit based on the original:

![Editing an image with GPT-Image-2](https://ofox.ai/screenshots/cherry-studio/new-04-image-edit.jpg)

## FAQ

**Q: The model list is empty after clicking "Manage"**

Check that the API Host is correct (no trailing slash) and that the API key was copied completely without extra spaces.

**Q: Connection failed after clicking "Check"**

1. Confirm the API Key was copied completely from the [OfoxAI Console](https://app.ofox.ai) without extra spaces
2. Confirm the API Host is correct
3. Confirm your network connection is working

**Q: The provider doesn't appear in the model selector**

Check that the toggle in the top-right corner of the configuration page is enabled.