Gemini Embedding 2
Multimodal embedding model mapping text, images, video, audio, and PDFs into a unified embedding space
GoogleEmbeddingCommon
Gemini Embedding 2
Multimodal embedding model mapping text, images, video, audio, and PDFs into a unified embedding space
gemini-embedding-2- Type
- Embedding
- Quality
- Not rated yet
- Context window
- 8,192 tokens
- Max output
- 1 tokens
- Input price
- $0.20 / 1M tokens
- Output price
- Not billed
- Cache read
- —
- Cache write
- —
- Released
- Apr 22, 2026
- Knowledge cutoff
- Nov 2025
- Takes in
- Text, Image, Audio, Video, PDF
- Puts out
- Text
- Reasoning
- No
- Tool calling
- No
- Structured output
- No
- File attachments
- Yes
- Access
- Paid
- License
- Proprietary
- Rarity
- Common
- Set number
- 885/1044
Weights aren’t public. Use is governed by Google’s terms. Specs from models.dev, last changed Oct 19, 2026. Official source: ai.google.dev.