Gemini API (embeddings

Convert text into high-dimensional vectors using Google’s text-embedding-004 model for semantic search and RAG workflows.

The Gemini API provides access to the text-embedding-004 model (optimized for 768-dimensional output) to power tasks like document retrieval, clustering, and classification. It supports a 2,048-token input window and includes a task_type parameter to refine vector generation for specific use cases (such as retrieval_query or semantic_similarity). Developers can integrate these embeddings with vector databases like Pinecone or Vertex AI Vector Search to build production-ready Retrieval-Augmented Generation (RAG) systems.

https://ai.google.dev/gemini-api/docs/embeddings