Compare models
Choose two or three models, including exact thinking and fast versions. Copy the URL to share your selection.
| Comparison | Gemini Embedding 2google/gemini-embedding-2 | Choose model 2 |
|---|---|---|
| Type | Embedding | — |
| Best for | Semantic search, RAG and clustering | — |
| TokenRex price | $0.20 per 1M tokens | — |
| Cached input / 1M tokens | No separate cached rate listed | — |
| Cache write / 1M tokens | Not applicable | — |
| Cached output | Not offered · standard output rate applies | — |
| Market price | $0.20 per 1M tokens | — |
| Input token limit (provider) | Within the context window | — |
| Vendor input modalities | Text, Image, Audio, Video, Documents | — |
| Vendor output modalities | Embeddings | — |
| TokenRex endpoints | POST /v1/embeddings | — |
| Benchmark evidence | 3 source-linked results | — |
| MTEB (Multilingual), Text-Text | 69.9mean task score Text-to-text evaluation only (model is natively multimodal but this metric isolates text-text performance). Published source text: Official Google performance table: Gemini Embedding 2 scores 69.9 Mean(Task) on MTEB (Multilingual), Text-Text, vs 68.4 for legacy gemini-embedding-001. Google DeepMind — Gemini Embedding 2 model card | — |
| MTEB (Code) | 84mean task score MTEB Code benchmark mean task score. Published source text: Official Google performance table: Gemini Embedding 2 scores 84.0 Mean(Task) on MTEB (Code) vs 76.0 for legacy gemini-embedding-001. Google DeepMind — Gemini Embedding 2 model card | — |
| Image-Text (TextCaps) | 97.4% Cross-modal image-to-text retrieval, TextCaps recall@1. Published source text: Official Google performance table: Gemini Embedding 2 Image→Text TextCaps recall@1 = 97.4 (vs 88.1 for legacy multimodalembedding@001, 88.9 Amazon Nova 2, 88.6 Voyage Multimodal 3.5). Google DeepMind — Gemini Embedding 2 model card | — |
No overall score, winner or cost-versus-score frontier is calculated without comparable, attributable benchmark evidence. Different test configurations appear separately. Vendor capabilities are distinct from TokenRex API support.
— means no published value or no matching capability. Provider token limits are not tested gateway guarantees; TokenRex media controls and response formats are listed separately.
Gemini Embedding 2
Sources and verification
Provider specifications describe the upstream model, not a guarantee of every feature through TokenRex. Prices come from the enabled catalog; benchmark results belong to the exact tested model.
- Vendor model documentation
Vendor-reported
Last checked - Google DeepMind — Gemini Embedding 2 model card
Vendor-reported
Last checked
Research notes (1)
- Vendor multimodal embeddings do not guarantee the same input surface through TokenRex.