Mistral

Direct access to Mistral's La Plateforme endpoints. Use it when your stack standardizes on Mistral for chat text with streaming, speech synthesis, transcription, and embeddings.

Capabilities

Text with streaming, audio generation, transcription, and embeddings.

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' => [
        'providers' => [
            'mistral' => [
                'className' => 'Crustum\Ai\Providers\MistralProvider',
                'key' => env('MISTRAL_API_KEY'),
                'url' => env('MISTRAL_URL', 'https://api.mistral.ai/v1'),
                'models' => [
                    'text' => [
                        'default' => env('MISTRAL_TEXT_MODEL', 'mistral-large-2512'),
                        'cheapest' => env('MISTRAL_TEXT_MODEL_CHEAPEST', 'mistral-small-4'),
                        'smartest' => env('MISTRAL_TEXT_MODEL_SMARTEST', 'mistral-medium-3-5'),
                    ],
                    'embeddings' => [
                        'default' => env('MISTRAL_EMBEDDINGS_MODEL', 'mistral-embed-2312'),
                        'dimensions' => (int)env('MISTRAL_EMBEDDINGS_DIMENSIONS', 1024),
                    ],
                    'transcription' => [
                        'default' => env('MISTRAL_TRANSCRIPTION_MODEL', 'voxtral-mini-2602'),
                    ],
                ],
            ],
        ],
    ],
];

Config key	Env var	Purpose
className	—	Provider class. Keep the default unless subclassing.
key (apiKey also accepted)	MISTRAL_API_KEY	Key from console.mistral.ai. Required.
url	MISTRAL_URL	API base URL. Change only for proxies or gateways.
models.text.default	MISTRAL_TEXT_MODEL	Chat model for normal calls.
models.text.cheapest	MISTRAL_TEXT_MODEL_CHEAPEST	Budget model.
models.text.smartest	MISTRAL_TEXT_MODEL_SMARTEST	Flagship model.
models.audio.default	MISTRAL_AUDIO_MODEL	Text-to-speech model (no env mapping in the default file; add one if you use audio).
models.transcription.default	MISTRAL_TRANSCRIPTION_MODEL	Speech-to-text (Voxtral) model.
models.embeddings.default	MISTRAL_EMBEDDINGS_MODEL	Embeddings model.
models.embeddings.dimensions	MISTRAL_EMBEDDINGS_DIMENSIONS	Vector width of the embeddings model.
headers	— (config only)	Extra HTTP headers merged into every request. Add as a plain array if needed.

Model selection priority per call: explicit model: argument wins, then the matching models.* value above, then the provider's built-in fallback.

See also

• Providers — capability matrix
• Audio — speech and transcription
• Embeddings — vectors and caching