> Portal Navigation:
> 
> - Append `.md` to any URL under `https://dev.wix.com/docs/` to get its markdown version.
> - Pages are either content pages (article or reference text) or menu pages (a list of links to child pages).
> - To get a menu page, truncate any URL to a parent path and append `.md` (e.g. `https://dev.wix.com/docs/sdk.md`, `https://dev.wix.com/docs/sdk/core-modules.md`).
> - Top-level index of all portals: https://dev.wix.com/docs/llms.txt
> - Full concatenated docs: https://dev.wix.com/docs/llms-full.txt

## Resource: About the Wix AI APIs

## Article: About the Wix AI APIs

## Article Link: https://dev.wix.com/docs/api-reference/articles/ai-tools/ai-apis/about-the-wix-ai-apis.md

## Article Content:

# About the Wix AI APIs

<blockquote class="caution">

<strong>Developer Preview</strong>

This API is subject to change. Bug fixes and new features will be released based on developer feedback throughout the preview period.

</blockquote>

> **Authentication and Permissions**
>
> To use the AI APIs, you must be authenticated as a Wix app or Wix user [identity](https://dev.wix.com/docs/api-reference/articles/authentication/about-identities.md). When [developing a site](https://dev.wix.com/docs/api-reference/articles/platform-overview/about-wix-site-development.md), you can also authenticate with an [API key](https://dev.wix.com/docs/api-reference/articles/authentication/api-keys/about-api-keys.md).
>
> For [app development](https://dev.wix.com/docs/build-apps/develop-your-app/about-developing-apps.md), you must also include the [`INVOKE AI MODELS`](https://manage.wix.com/app-selector?title=Select+an+App&subtitle=After+Selection+you+will+be+redirected+to+App+Permissions+Page&primaryButtonText=Go+to+permissions+page&actionUrl=https%3A%2F%2Fdev.wix.com%2Fapps%2F%7BappId%7D%2Fdev-center-permissions%2Fadd%3FselectedPermission%3DSCOPE.DATA_SCIENCE.INVOKE_AI_MODELS%26filterByPermissionName%3DSCOPE.DATA_SCIENCE.INVOKE_AI_MODELS) permission scope. Learn more about [app permissions](https://dev.wix.com/docs/build-apps/develop-your-app/access/authorization/about-permissions.md).

::::tabs
:::REST_TAB
The Wix AI APIs provide a streamlined way to interact with multiple AI providers in your code while Wix handles authentication and billing:
:::
:::SDK_TAB
Built on the [Vercel AI SDK](https://ai-sdk.dev/), the Wix AI APIs provide a streamlined way to interact with multiple AI providers in your code while Wix handles authentication and billing:
:::
::::

- When [developing a site](https://dev.wix.com/docs/api-reference/articles/platform-overview/about-wix-site-development.md), Wix bills based on AI model usage by you, your collaborators, members, and visitors.
- When [building a Wix app](https://dev.wix.com/docs/build-apps/get-started/overview/about-wix-apps.md), Wix bills the Wix users who install your app, so you don't need to set up billing integration.

::::tabs
:::REST_TAB

:::
:::SDK_TAB

Learn how to [set up the Wix AI SDK](https://dev.wix.com/docs/api-reference/articles/ai-tools/ai-apis/set-up-the-wix-ai-sdk.md).
:::
::::

<blockquote class="important">

**Important:**
Each method call uses approximately 1 AI credit. Learn more about [AI credits on Wix](https://support.wix.com/en/article/about-ai-credits).

</blockquote>

## Providers and models

Wix provides access to models from the following providers. To call any supported provider method, use this base URL pattern:

```curl
https://www.wixapis.com/{provider}/v1/{method-path}
```

To view the full list of models, call each provider's method at the specified URL. See each provider's documentation for request parameters and response details:

| Provider  | Method                                                                    | URL to call                                   |
| --------- | ------------------------------------------------------------------------- | --------------------------------------------- |
| OpenAI    | [List Models](https://platform.openai.com/docs/api-reference/models/list) | `https://www.wixapis.com/openai/v1/models`    |
| Anthropic | [List Models](https://platform.claude.com/docs/en/api/models/list)        | `https://www.wixapis.com/anthropic/v1/models` |

> **Note:** When a provider releases a new model, it might take a few days before Wix supports it.

## Generating text

::::tabs
:::REST_TAB

Generate text from prompts using models by the following providers:

### OpenAI

OpenAI methods are available on the following base URL:

```curl
https://www.wixapis.com/openai/v1/{method-path}
```

Generate text by calling any supported OpenAI method on its path:

| Name                                                                                       | Description                                                | Path                |
| ------------------------------------------------------------------------------------------ | ---------------------------------------------------------- | ------------------- |
| [Create a model response](https://platform.openai.com/docs/api-reference/responses/create) | Create a response from text or image prompts.              | `/responses`        |
| [Create chat completion](https://platform.openai.com/docs/api-reference/chat/create)       | Create a model response for a specified chat conversation. | `/chat/completions` |

> **Note:** To stream the response, set the `stream` parameter to `true`.

For example, this request uses the GPT-5 model to generate a response to the prompt "Tell me a bedtime story about a unicorn":

```curl
curl -X POST 'https://www.wixapis.com/openai/v1/responses' \
-H 'Content-Type: application/json' \
-H 'Authorization: <AUTH>' \
-d '{
  "model": "gpt-5",
  "input": "Tell me a bedtime story about a unicorn."
}'
```

### Anthropic

Anthropic methods are available on the following base URL:

```curl
https://www.wixapis.com/anthropic/v1/{method-path}
```

Generate text by calling any supported Anthropic method on its path:

| Name                                                                        | Description                                     | Path        |
| --------------------------------------------------------------------------- | ----------------------------------------------- | ----------- |
| [Create a message](https://platform.claude.com/docs/en/api/messages/create) | Create a message from a list of input messages. | `/messages` |

> **Note:** To stream the response, set the `stream` parameter to `true`.

For example, this request uses the Claude Opus 4.6 model to stream a response to the prompt "Hello, world":

```curl
curl -X POST 'https://www.wixapis.com/anthropic/v1/messages' \
-H 'Content-Type: application/json' \
-H 'Authorization: <AUTH>' \
-d '{
  "messages": [
    {
      "content": "Hello, world",
      "role": "user"
    }
  ],
  "model": "claude-opus-4-6",
  "stream": true
}'
```

:::
:::SDK_TAB

Generate text from prompts with the following methods:

- [`generateText()`](https://ai-sdk.dev/docs/reference/ai-sdk-core/generate-text): Generates text for a specified prompt and model.
- [`streamText()`](https://ai-sdk.dev/docs/reference/ai-sdk-core/stream-text): Streams text from a specified prompt and model.

For example, this code snippet generates text using OpenAI's GPT-5 model via the Responses API:

```js
import { generateText } from "ai";
import { openai } from "@wix/ai";

const { text } = await generateText({
  model: openai.responses("gpt-5"),
  prompt: "Invent a new holiday and describe its traditions.",
});

console.log(text);
```

This code snippet generates text using Anthropic's Claude Opus 4.6 model:

```js
import { generateText } from "ai";
import { anthropic } from "@wix/ai";

const { text } = await generateText({
  model: anthropic("claude-opus-4-6"),
  prompt: "Tell me a bedtime story about a unicorn.",
});

console.log(text);
```

:::
::::

### Suggested models

You can generate text using these suggested models from the following providers:

| Provider  | Model               |
| --------- | ------------------- |
| OpenAI    | `gpt-5.2-pro`       |
| OpenAI    | `gpt-5.2`           |
| OpenAI    | `gpt-5.1`           |
| OpenAI    | `gpt-5-nano`        |
| OpenAI    | `gpt-5-mini`        |
| OpenAI    | `gpt-5`             |
| OpenAI    | `gpt-4.1-nano`      |
| OpenAI    | `gpt-4.1-mini`      |
| OpenAI    | `gpt-4.1`           |
| Anthropic | `claude-opus-4-7`   |
| Anthropic | `claude-sonnet-4-6` |
| Anthropic | `claude-haiku-4-5`  |

Learn more about [providers and models](#providers-and-models).

## Generating embeddings

::::tabs
:::REST_TAB

Create vector representations for semantic comparisons and search using models by the following providers:

### OpenAI

OpenAI methods are available on the following base URL:

```curl
https://www.wixapis.com/openai/v1/{method-path}
```

Generate embeddings by calling the supported OpenAI method on its path:

| Name                                                                                  | Description                                              | Path          |
| ------------------------------------------------------------------------------------- | -------------------------------------------------------- | ------------- |
| [Create embeddings](https://platform.openai.com/docs/api-reference/embeddings/create) | Create an embeddings vector representing the input text. | `/embeddings` |

For example, this request uses the text-embedding-ada-002 model to create an embedding vector:

```curl
curl -X POST 'https://www.wixapis.com/openai/v1/embeddings' \
-H 'Content-Type: application/json' \
-H 'Authorization: <AUTH>' \
-d '{
  "input": "The food was delicious but the waiter was rude.",
  "model": "text-embedding-ada-002",
  "encoding_format": "float"
}'
```

:::
:::SDK_TAB

Create vector representations for semantic comparisons and search by calling the following methods:

- [`embed()`](https://ai-sdk.dev/docs/reference/ai-sdk-core/embed): Embed a single value.
- [`embedMany()`](https://ai-sdk.dev/docs/reference/ai-sdk-core/embed-many): Embed multiple values at once.

For example, this code snippet uses OpenAI's text-embedding-3-small model via the [Embeddings API](https://ai-sdk.dev/docs/ai-sdk-core/embeddings) to create multiple embedding vectors:

```js
import { embedMany } from "ai";
import { openai } from "@wix/ai";

const { embeddings } = await embedMany({
  model: openai.embeddingModel("text-embedding-3-small"),
  values: [
    "sunny day at the beach",
    "rainy afternoon in the city",
    "snowy night in the mountains",
  ],
});
```

:::
::::

### Suggested models

You can create embeddings using these suggested models from the following providers:

| Provider | Model                    |
| -------- | ------------------------ |
| OpenAI   | `text-embedding-3-small` |
| OpenAI   | `text-embedding-3-large` |
| OpenAI   | `text-embedding-ada-002` |

Learn more about [providers and models](#providers-and-models).

## Generating images

::::tabs
:::REST_TAB

Generate images from text prompts using models by the following providers:

### Runware

Access [Runware](https://runware.ai/docs/image-inference/introduction) image generation functionality using the following URL:

```curl
https://www.wixapis.com/runwareschemaless/v1/request
```

To interact with Runware models, send a [request with an array of tasks](https://runware.ai/docs/image-inference/api-reference#request), each containing the required configuration details.

For example, this request generates an image using Google's Nano Banana Pro 2 model:

```curl
curl -X POST 'https://www.wixapis.com/runwareschemaless/v1/request' \
-H 'Content-Type: application/json' \
-H 'Authorization: <AUTH>' \
-d '[{
  "taskType": "imageInference",
  "taskUUID": "a770f077-f413-47de-9dac-be0b26a35da6",
  "outputType": "URL",
  "outputFormat": "jpg",
  "positivePrompt": "a serene mountain landscape with a crystal-clear lake reflecting the sky",
  "height": 1024,
  "width": 1024,
  "model": "google:4@2",
  "steps": 30,
  "CFGScale": 7.5,
  "numberResults": 4
}]'
```

:::
:::SDK_TAB

Generate images from text prompts by calling the following method:

- [`generateImage()`](https://ai-sdk.dev/docs/reference/ai-sdk-core/generate-image)

For example, this code snippet uses Google's Nano Banana Pro 2 model to generate an image:

```js
import { generateImage } from "ai";
import { runware } from "@wix/ai";

const result = await generateImage({
  model: runware.image("google:4@2"),
  prompt: "A futuristic cityscape at night with neon lights",
});
```

:::
::::

### Suggested models

You can generate images using these suggested [Runware models](https://runware.ai/models). Specify the [model's AIR ID](https://runware.ai/docs/image-inference/models#air-system) in the image generation method:

| Model name         | Model AIR ID             |
| ------------------ | ------------------------ |
| FLUX.2 Dev         | `runware:400@1`          |
| FLUX.2 Pro         | `bfl:5@1`                |
| Ideogram 3.0       | `ideogram:4@1`           |
| ImagineArt 1.5 Pro | `imagineart:1.5-pro@0`   |
| Nano Banana Pro 2  | `google:4@2`             |
| Seedream 4.5       | `bytedance:seedream@4.5` |
| Z-Image-Turbo      | `runware:z-image@turbo`  |

> **Note:** You can use any model supported by Runware. [Sign up to Runware](https://my.runware.ai/signup) to explore the full model list.

## See also

- [Set up the Wix AI SDK](https://dev.wix.com/docs/api-reference/articles/ai-tools/ai-apis/set-up-the-wix-ai-sdk.md)
- [About AI credits on Wix](https://support.wix.com/en/article/about-ai-credits)
- [Vercel AI SDK Core](https://ai-sdk.dev/docs/ai-sdk-core/overview)