TypeSafe

A classification-only provider: no text, no streams, no media. Use it to get typed answers with probabilities — you define the answer shape (an enum or compatible type) and the model picks the most likely value plus a confidence score. It is the default backend behind default_for_classification.

Capabilities

Classification.

Configuration

Configure via environment variables. config/ai.php only forwards them with env() — set the values in .env or the host environment, never hardcode keys in the config file.

// In config/ai.php
return [
    'Ai' => [
        'default_for_classification' => env('AI_DEFAULT_FOR_CLASSIFICATION', 'typesafe'),
        'providers' => [
            'typesafe' => [
                'className' => 'Crustum\Ai\Providers\TypeSafeProvider',
                'key' => env('TYPESAFE_API_KEY'),
                'url' => env('TYPESAFE_URL', 'https://api.typesafe.ai/v1'),
                'models' => [
                    'classification' => [
                        'default' => env('TYPESAFE_CLASSIFICATION_MODEL', 'jev-latest'),
                    ],
                ],
            ],
        ],
    ],
];

Config key	Env var	Purpose
className	—	Provider class. Keep the default unless subclassing.
key (apiKey also accepted)	TYPESAFE_API_KEY	TypeSafe key. Required.
url	TYPESAFE_URL	API base URL. Change only for proxies or gateways.
models.classification.default	TYPESAFE_CLASSIFICATION_MODEL	Classification model answering typed questions with probabilities.
headers	— (config only)	Extra HTTP headers merged into every request. Add as a plain array if needed.

Per-call model: overrides the configured model for that call.

See also

• Providers — capability matrix
• Classification — typed answers with probabilities