Gemini Embedding 2 Preview

Gemini Embeddings is a multimodal embedding technique that converts text, audio, video, and image data into numerical vectors that can be processed by machine learning algorithms, especially large models. These vector representations are designed to capture the semantic meaning and context of the data they represent. 通过统一 API 即可完成接入，支持弹性并发、用量监控和团队级密钥管理，适合从快速验证到稳定调用的完整链路。