{"id":122309,"date":"2026-03-24T21:18:25","date_gmt":"2026-03-24T21:18:25","guid":{"rendered":"https:\/\/make.wordpress.org\/core\/?p=122309"},"modified":"2026-03-24T21:18:25","modified_gmt":"2026-03-24T21:18:25","slug":"introducing-the-ai-client-in-wordpress-7-0","status":"publish","type":"post","link":"https:\/\/make.wordpress.org\/core\/2026\/03\/24\/introducing-the-ai-client-in-wordpress-7-0\/","title":{"rendered":"Introducing the AI Client in WordPress 7.0"},"content":{"rendered":"

WordPress 7.0 includes a built-in AI Client \u2014 a provider-agnostic PHPPHP<\/span> The web scripting language in which WordPress is primarily architected. WordPress requires PHP 7.4 or higher<\/span><\/span><\/span> APIAPI<\/span> An API or Application Programming Interface is a software intermediary that allows programs to interact with each other and share data in limited, clearly defined ways.<\/span><\/span><\/span> that lets plugins send prompts to AI models and receive results through a consistent interface. Your pluginPlugin<\/span> A plugin is a piece of software containing a group of functions that can be added to a WordPress website. They can extend functionality or add new features to your WordPress websites. WordPress plugins are written in the PHP programming language and integrate seamlessly with WordPress. These can be free in the WordPress.org Plugin Directory https:\/\/wordpress.org\/plugins\/<\/a> or can be cost-based plugin from a third-party.<\/span><\/span><\/span> describes what<\/em> it needs and how<\/em> it needs it. WordPress handles routing the request to a suitable model from a provider the site owner has configured.<\/p>\n\n\n\n

This post explains the API surface, walks through code examples, and covers what plugin developers need to know.<\/p>\n\n\n\n

The entry point: wp_ai_client_prompt()<\/code><\/h2>\n\n\n\n

Every interaction starts with:<\/p>\n\n\n\n

$builder = wp_ai_client_prompt();<\/code><\/pre>\n\n\n\n
This returns a WP_AI_Client_Prompt_Builder<\/code> object, a fluent builder that offers a myriad of ways to customize your prompt. You chain configuration methods and then call a generation method to receive a result:<\/p>\n\n\n\n
$text = wp_ai_client_prompt( 'Summarize the benefits of caching in WordPress.' )\n    ->using_temperature( 0.7 )\n    ->generate_text();<\/code><\/pre>\n\n\n\n
You can pass the prompt text directly as a parameter to wp_ai_client_prompt()<\/code> for convenience, though alternatively the with_text()<\/code> method is available for building the prompt incrementally.<\/p>\n\n\n\n
Text generation<\/h2>\n\n\n\n
Here\u2019s a basic text generation example:<\/p>\n\n\n\n
$text = wp_ai_client_prompt( 'Write a haiku about WordPress.' )\n    ->generate_text();\n\nif ( is_wp_error( $text ) ) {\n    \/\/ Handle error.\n    return;\n}\n\necho wp_kses_post( $text );<\/code><\/pre>\n\n\n\n
You can pass a JSONJSON<\/span> JSON, or JavaScript Object Notation, is a minimal, readable format for structuring data. It is used primarily to transmit data between a server and web application, as an alternative to XML.<\/span><\/span><\/span> schema so that the model returns structured data as a JSON string:<\/p>\n\n\n\n
$schema = array(\n    'type'  => 'array',\n    'items' => array(\n        'type'       => 'object',\n        'properties' => array(\n            'plugin_name' => array( 'type' => 'string' ),\n            'category'    => array( 'type' => 'string' ),\n        ),\n        'required' => array( 'plugin_name', 'category' ),\n    ),\n);\n\n$json = wp_ai_client_prompt( 'List 5 popular WordPress plugins with their primary category.' )\n    ->as_json_response( $schema )\n    ->generate_text();\n\nif ( is_wp_error( $json ) ) {\n    \/\/ Handle error.\n    return;\n}\n\n$data = json_decode( $json, true );<\/code><\/pre>\n\n\n\n
You can request multiple response candidates as variations for the same prompt:<\/p>\n\n\n\n
$texts = wp_ai_client_prompt( 'Write a tagline for a photography blog.' )\n    ->generate_texts( 4 );<\/code><\/pre>\n\n\n\n
Image generation<\/h2>\n\n\n\n
Here\u2019s a basic image generation example:<\/p>\n\n\n\n
use WordPress\\AiClient\\Files\\DTO\\File;\n\n$image_file = wp_ai_client_prompt( 'A futuristic WordPress logo in neon style' )\n    ->generate_image();\n\nif ( is_wp_error( $image_file ) ) {\n    \/\/ Handle error.\n    return;\n}\n\necho '<img src=\"' . esc_url( $image_file->getDataUri() ) . '\" alt=\"\">';<\/code><\/pre>\n\n\n\n
generate_image()<\/code> returns a File<\/code> DTO with access to the image data via getDataUri()<\/code>.<\/p>\n\n\n\n
Similar to text generation, you can request multiple variations of the same image:<\/p>\n\n\n\n
$images = wp_ai_client_prompt( 'Aerial shot of snowy plains, cinematic.' )\n    ->generate_images( 4 );\n\nif ( is_wp_error( $images ) ) {\n    \/\/ Handle error.\n    return;\n}\n\nforeach ( $images as $image_file ) {\n    echo '<img src=\"' . esc_url( $image_file->getDataUri() ) . '\">';\n}<\/code><\/pre>\n\n\n\n
Getting the full result object<\/h2>\n\n\n\n
For richer metadata, e.g. covering provider and model information, use generate_*_result()<\/code> instead. For example, for image generation:<\/p>\n\n\n\n
$result = wp_ai_client_prompt( 'A serene mountain landscape.' )\n    ->generate_image_result();<\/code><\/pre>\n\n\n\n
This returns a GenerativeAiResult<\/code> object that provides several pieces of additional information, including token usage and which provider and which model responded to the prompt. The most relevant methods for this additional metadata are:<\/p>\n\n\n\n