Crédito gratuito · Uma chave · Pague conforme o uso
Endpoints Jina AI API
Experimentar uma tarefa
3 correspondentes a endpoints · 3 no totalComo funciona a cobrança
Os preços estão em USD. Uma chamada corresponde à execução de um endpoint (um Run). Uma tarefa pode usar várias chamadas.
“Tarifa de catálogo” é o preço listado por chamada; os totais baseados no uso dependem das entradas e opções. “Até” é a cobrança total máxima dessa chamada. As tarifas de resultados e tokens usam as unidades indicadas.
Abra um endpoint para consultar suas regras de cobrança, incluindo cobranças por nenhuma correspondência ou por erros.
Token top-ups for hosted APIs. Embeddings and reranking count input tokens. Top-ups fund the Jina API key’s token balance.
Uso e limites
Separate free and paid rate limits
Jina enforces requests-per-minute and tokens-per-minute limits by API key; limits vary by product and tier.
Contas e chaves
Jina API key
Direct embedding and reranking calls use your Jina API key with bearer authentication.
Configuração do agente
Jina REST API
Send text to the embedding endpoint or a query and documents to the reranking endpoint using Jina credentials.
Taxas de assinatura, saldos pré-pagos e tarifas de uso são exibidos separadamente por modelo de cobrança. O custo real da tarefa depende dos endpoints, das entradas e do volume.
Jina AI API perguntas respondidas.
Which Jina AI API operations are available through Glasser?
Glasser lists document reranking and two text embedding operations: jina-embeddings-v3 and jina-embeddings-v5-text-small. Both embedding operations use native late chunking.
What is the difference between embeddings and reranking?
Embeddings convert supplied text into vectors your application can index for similarity search. Reranking takes a query and supplied documents, then scores their relevance. Your application can use reranking after retrieving a candidate set.
How does native late chunking affect the input?
The two listed embedding operations have late_chunking enabled. All texts supplied in a request share the model’s context window, and the provider rejects input that exceeds it. Inspect the selected endpoint’s current input requirements before preparing a batch.
Can these Jina operations fetch websites or search the web?
These three operations process text or documents you supply. For source discovery or webpage extraction, choose a search or extraction provider first, then pass the text to the relevant Jina operation in your own workflow.
Does Glasser store the embeddings in a vector database?
The listed embedding operations generate vectors. Your application handles vector storage, indexing and retrieval. The reranker separately scores supplied documents against a query.
Do I need a Jina API key or a token top-up?
Calls through Glasser use your Glasser key and workspace balance. Direct Jina access uses a Jina API key and token balance, with separate free and paid rate limits. Jina offers token top-ups for its hosted APIs.
How much do Jina embedding and reranking calls cost?
Use the live Glasser endpoint directory for the current quote, billing unit and cap, then inspect the selected operation for its full price. Direct Jina embedding and reranking usage is counted from input tokens; the direct dashboard manages its token balance and top-ups.
