Description
EmpirioLabs AI is specialized AI model hosting for open, proprietary, and custom stacks. We run open-source models on GPU infrastructure we manage, serve optimized endpoints for proprietary models, and help teams deploy their own models to real users. The catalog spans text, image, video, audio, transcription, embedding, reranking, research, and 3D models, with a browser playground, GPU Cloud instances, and hosted AI agents.
This plugin registers EmpirioLabs AI with the AI Client built into WordPress, so any plugin or feature that uses the AI Client can run EmpirioLabs AI models. Add your API key on the Settings > Connectors screen and EmpirioLabs AI appears with the other AI providers on your site.
Features
- Text generation with chat models, including system instructions, multi-turn conversations, function calling, and JSON output on the models that support them.
- Image and audio input for the models that accept them.
- Image generation, with aspect ratio, orientation, number of images, and output format on the models that offer them.
- Text to speech, with a voice and an audio format on the models that offer them.
- Video generation from a description, with a source image given as a link on the models that take one, and with aspect ratio and orientation on the models that offer them. A video can be started in one request and collected in a later one.
- Embedding generation on WordPress versions whose AI Client includes embeddings (WordPress 7.2 and later).
- The model list loads from the EmpirioLabs AI catalog, so new models are available without a plugin update.
- The API key can come from the Settings > Connectors screen, the
EMPIRIOLABS_API_KEYenvironment variable, or a PHP constant of the same name.
Using EmpirioLabs AI in your own code
Any prompt sent through the AI Client can run on EmpirioLabs AI. To require it for a prompt:
$text = wp_ai_client_prompt( 'Write a two-sentence welcome message for a bakery website.' )
->using_provider( 'empiriolabs' )
->generate_text();
To prefer a specific model, pass its model ID, for example using_model_preference( 'MODEL_ID' ). Model IDs are listed at empiriolabs.ai/models. Without a preference, the AI Client picks the first EmpirioLabs AI model that supports the request.
To convert text to speech:
$audio = wp_ai_client_prompt( 'Welcome to our bakery. We open at seven every morning.' )
->using_provider( 'empiriolabs' )
->as_output_mime_type( 'audio/mpeg' )
->convert_text_to_speech();
To choose a voice, add as_output_speech_voice( 'VOICE' ) with one of the model’s voices. Each model’s page in the documentation lists its voices.
A video takes minutes to generate. Start it, keep its operation ID, and read the result in a later request:
$model = \WordPress\AiClient\AiClient::defaultRegistry()->getProviderModel( 'empiriolabs', 'MODEL_ID' );
$operation = $model->generateVideoOperation(
array(
new \WordPress\AiClient\Messages\DTO\UserMessage(
array( new \WordPress\AiClient\Messages\DTO\MessagePart( 'A paper boat drifting down a rainy street' ) )
),
)
);
$operation_id = $operation->getId();
// Later, with the saved operation ID:
$operation = \EmpirioLabs\AiProvider\Provider\EmpirioLabsProvider::operationsHandler()->getOperation( $operation_id );
if ( $operation->getState()->isSucceeded() ) {
$video_url = $operation->getResult()->toVideoFile()->getUrl();
}
External services
This plugin connects to the EmpirioLabs AI API at https://api.empiriolabs.ai to list and run models. The service is required for every feature of the plugin, and the plugin sends nothing until an API key is set.
What is sent, and when:
- When WordPress checks the connection, for example when you save a key on the Settings > Connectors screen or when a request needs a model, the plugin requests the list of models with your API key.
- When a plugin or feature on your site generates text, images, speech, video, or embeddings through the AI Client with EmpirioLabs AI, the plugin sends the content of that request (the prompt, earlier messages of the conversation, the system instruction, any attached images or audio, and the generation settings), the model ID, and your API key.
- While it waits for a video, and when your code reads a video that was started earlier, the plugin requests the state of that generation with its operation ID and your API key.
- Every request identifies the plugin by its name and version in the User-Agent header.
The service is provided by EmpirioLabs.ai LLC: Terms of Service, Privacy Policy.
Screenshots


Installation
- Install the plugin from Plugins > Add New, or upload the
empiriolabs-ai-providerfolder to the/wp-content/plugins/directory. - Activate the plugin on the Plugins screen.
- Create an API key on the API Keys page of the EmpirioLabs AI dashboard.
- Go to Settings > Connectors and paste the key into the EmpirioLabs AI connector.
To set the key outside the database, define EMPIRIOLABS_API_KEY as an environment variable or as a constant in wp-config.php:
define( 'EMPIRIOLABS_API_KEY', 'your-api-key' );
FAQ
-
How do I get an API key?
-
Sign in or create an account at platform.empiriolabs.ai, then create a key on the API Keys page. Requests use the credits on your EmpirioLabs AI account at the prices on the pricing page.
-
Which models can I use?
-
The text, image, speech, video, and embedding models in the EmpirioLabs AI catalog that the AI Client can call. The plugin reads the model list from the API, so new models appear without a plugin update. Browse the models at empiriolabs.ai/models.
-
Does the plugin need WordPress 7.0?
-
Yes. The plugin adds a provider to the AI Client that ships with WordPress 7.0 and later.
-
An image request fails after 30 seconds
-
WordPress allows each AI Client request 30 seconds by default, and some image models take longer. Raise the limit with the
wp_ai_client_default_request_timeoutfilter, for example to 120 seconds:add_filter( 'wp_ai_client_default_request_timeout', function () { return 120; } ); -
A video request ends before the video is ready
-
A video usually takes several minutes.
generate_video()waits for it for as long as the request timeout allows, 30 seconds by default (the previous question shows how to raise it). A video that needs longer keeps running, and the error message contains its operation ID. To avoid waiting, start the video withgenerateVideoOperation()and read the operation in a later request, as shown in the description. -
Where is my API key stored?
-
A key saved on the Settings > Connectors screen is stored in your site’s database, and WordPress shows only its last four characters afterward. Deleting the plugin removes the saved key. A key set with the
EMPIRIOLABS_API_KEYenvironment variable or constant is never stored in the database. -
Where can I get help?
-
Read the EmpirioLabs AI documentation or email support@empiriolabs.ai.
Reviews
There are no reviews for this plugin.
Contributors & Developers
“EmpirioLabs AI Provider” is open source software. The following people have contributed to this plugin.
ContributorsTranslate “EmpirioLabs AI Provider” into your language.
Interested in development?
Browse the code, check out the SVN repository, or subscribe to the development log by RSS.
Changelog
1.1.0
- Text to speech with the speech models of the catalog, with a voice and an audio format on the models that offer them.
- Video generation with the video models of the catalog, from a description and an optional source image link. A video can be started in one request and read in a later one with its operation ID.
- The model list is read again right after a plugin update.
1.0.0
- Initial release: text generation, image generation, and embedding generation with EmpirioLabs AI models through the WordPress AI Client.
