# Create Model

Create new model.
Compatibility Rules:  FIBO Models: Must use a dataset with training_version = fibo.  Legacy Models: Must use a dataset with training_version in [max, light, 3.2, 2.3].  Cross-Compatibility:  Legacy datasets are generally interchangeable (e.g., a max dataset can train a light model). * FIBO is isolated. You cannot train a legacy model on a FIBO dataset or vice versa.

Endpoint: POST /tailored-gen/models

## Header parameters:

  - `api_token` (string, required)

## Request fields (application/json):

  - `name` (string)
    Name of the model (required)

  - `dataset_id` (integer)
    ID of the dataset to use (required)

  - `training_mode` (string)
    Defines the training configuration strategy.
* fully_automated: Get good results without the guesswork. Automatically configures the training recipe based on dataset size and IP medium/type.
* expert: Designed for experienced users needing control over training parameters and support for larger datasets.
    Enum: "fully_automated", "expert"

  - `training_version` (string)
    Defines the foundational model used for fine-tuning.
Must be compatible with the dataset's training_version.
    Enum: "light", "max", "bria-2.3", "bria-3.2", "fibo"

  - `description` (string)
    Description of the model (optional)

## Response 201 fields (application/json):

  - `id` (integer)
    Unique identifier for the model

  - `name` (string)
    Name of the model

  - `description` (string)
    Description of the model

  - `status` (string)
    Status of the model
    Enum: "Created"

  - `training_version` (string)
    Training version
    Enum: "light", "max", "bria-2.3", "bria-3.2"

  - `training_mode` (string)
    Training configuration strategy
    Enum: "fully_automated", "expert"

  - `generation_prefix` (string,null)
    Text automatically prepended to prompts.
* For fibo models: Null.
* For legacy models: String matching the dataset's caption_prefix.

  - `project_id` (integer)
    ID of the project this model belongs to

  - `dataset_id` (integer)
    ID of the dataset used for training

  - `created_at` (string)
    Timestamp when the model was created

  - `updated_at` (string)
    Timestamp when the model was last updated.

  - `custom_metadata` (string)
    Custom metadata associated with the model.

  - `active_model_version` (string)
    The active version (checkpoint) of the model. becomes relevant after a training has been completed.

  - `training_parameters` (object)
    The training parameters used for the model. will be filled using the /start_training endpoint.

  - `checkpoint_steps` (array)
    A list of checkpoints saved for the model. will be filled using the /start_training endpoint.

## Response 400 fields (application/json):

  - `error_code` (string)

  - `message` (string)


## Response 401 fields

## Response 404 fields

## Response 429 fields

## Response 500 fields