Gemini Embedding
Google's embedding model that maps text (and, in multimodal variants, images, video, audio, and PDFs) into a shared vector space, leading retrieval benchmarks (~67.7 MTEB retrieval). It supports Matryoshka dimensions up to 3,072 and integrates with the Gemini API and Vertex AI.
Strengths
- Leads MTEB retrieval benchmarks (~67.7)
- Shared multimodal vector space (text, image, and more)
- Matryoshka dimensions up to 3,072
- Gemini API and Vertex AI integration
- Strong multilingual retrieval
Caveats
- Proprietary - Google API only (no self-hosting)
- Per-request input token limits
- Usage-based pricing scales with volume
- Vendor
- Context Window
- 2,048 tokens
- License
- Proprietary
- Modalities
- textimage
Capabilities
Vision
Audio
Video
Tool Use
Resources
Pricing
Usage-based via the Gemini API and Vertex AI (per 1M tokens); see Google pricing for current rates.