Jina

Jina's search stack for the retrieval half of RAG: embeddings plus reranking. Use it when the job is text in, vectors or orderings out, and text generation lives on another provider.

Capabilities

Embeddings and reranking.

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' => [
            'jina' => [
                'className' => 'Crustum\Ai\Providers\JinaProvider',
                'key' => env('JINA_API_KEY'),
                'url' => env('JINA_URL', 'https://api.jina.ai/v1'),
                'models' => [
                    'embeddings' => [
                        'default' => env('JINA_EMBEDDINGS_MODEL', 'jina-embeddings-v4'),
                        'dimensions' => (int)env('JINA_EMBEDDINGS_DIMENSIONS', 2048),
                    ],
                    'reranking' => [
                        'default' => env('JINA_RERANKING_MODEL', 'jina-reranker-v3.5'),
                    ],
                ],
            ],
        ],
    ],
];

Config key	Env var	Purpose
className	—	Provider class. Keep the default unless subclassing.
key (apiKey also accepted)	JINA_API_KEY	Key from jina.ai. Required.
url	JINA_URL	API base URL. Change only for proxies or gateways.
models.embeddings.default	JINA_EMBEDDINGS_MODEL	Embeddings model.
models.embeddings.dimensions	JINA_EMBEDDINGS_DIMENSIONS	Vector width. Must match the embeddings model.
models.reranking.default	JINA_RERANKING_MODEL	Reranker used when no per-call model is given.
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
• Embeddings — vectors and caching
• Reranking — ordering search results