Retrieve the list of available base models.
This endpoint returns all general-purpose base models available for use in chat completions, embeddings, and domain alignment workflows.
Returns:
models: List of base model objects, each containing:
id: Unique model identifiername: Human-readable model nametype: Model typedescription: Model description and capabilitiesparameters: Model parameter countcapabilities: Supported features (e.g., chat, embeddings, vision)context_window: Maximum context length in tokenstraining_data_cutoff: Training data cutoff dateimplementation_type: Implementation typeendpoint: API endpointalignment_ready: Whether model can be used for alignmentis_active: Whether model is currently activenum_params_in_B: Number of parameters in billionsExample Request:
GET /api/v3/models/base
Headers: {"Authorization": "Bearer <api_key>"}
Example Response:
{
"models": [
{
"id": "nugen-flash-instruct",
"name": "Nugen Flash Instruct",
"type": "text-generation",
"description": "Fast instruction-following model",
"endpoint": null,
"alignment_ready": true,
"is_active": true,
"num_params_in_B": 7.0
},
{
"id": "model-20240229",
"name": "test-model",
"type": "text-generation",
"description": "Balanced intelligence and speed",
"endpoint": null,
"alignment_ready": false,
"is_active": true,
"num_params_in_B": 200.0
}
]
}
Notes:
/models/aligned to see your custom domain-aligned modelsalignment_ready indicates if the model can be used as a base for domain alignmentBearer authentication header of the form Bearer <token>, where <token> is your auth token.
Returns a list of all available base models that can be used for domain alignment. Each model object includes details such as model ID, name, type, description, parameter count, capabilities, context window size, training data cutoff date, implementation type, and whether it is ready for alignment. Use this endpoint to explore the general-purpose models you can use as a starting point for creating custom aligned models.