Gemini Embedding 2

多模态向量模型，将文本、图像、视频、音频和 PDF 映射到同一向量空间。

Googlegemini-embedding-22026-04-22报告数据错误

规格

上下文: —
最大输出: —
输入价: $0.2/ 百万 Token
输出价: —
发布日期: 2026-04-22

上下文

上下文: —
最大输入: 8.2K
最大输出: —

价格 · USD / 百万 Token

输入: $0.2
输出: —
缓存读取: —
缓存写入: —

输入输出

输入: 文本图像视频音频文档
输出: 向量

API

接口类型: embeddings

思考

支持思考: 不支持

源信息

状态: 可用
发布日期: 2026-04-22
知识截止: —

能力

官方未公布

调用方式

googlemodel = gemini-embedding-2

embeddings

POSThttps://generativelanguage.googleapis.com/v1beta/models/{model}:embedContent

JSON

gemini-embedding-2.json

{
  "id": "gemini-embedding-2",
  "object": "model",
  "created": 1776816000,
  "owned_by": "google",
  "name": "Gemini Embedding 2",
  "api": {
    "types": [
      "embeddings"
    ]
  },
  "limits": {
    "context": null,
    "input": 8192,
    "output": null
  },
  "modalities": {
    "input": [
      "text",
      "image",
      "video",
      "audio",
      "document"
    ],
    "output": [
      "embedding"
    ]
  },
  "reasoning": {
    "supported": false,
    "efforts": []
  },
  "pricing": {
    "currency": "USD",
    "unit": "1M_tokens",
    "input": 0.2,
    "output": null,
    "cache_read": null,
    "cache_write": null
  },
  "features": [],
  "info": {
    "status": "active",
    "release_date": "2026-04-22",
    "knowledge_cutoff": null,
    "description": "Multimodal embedding model mapping text, images, video, audio and PDFs into one embedding space.",
    "docs": "https://ai.google.dev/gemini-api/docs/models/gemini-embedding-2",
    "verified_at": "2026-10-11"
  }
}

官方来源