Gemini Embedding 001 vs Gemini Embedding 2
MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) Embeddings assign examples to categories, group similar examples, estimate semantic similarity between pairs, classify relationships between pairs, rank relevant documents for queries, find translated sentence pairs, rerank documents according to instructions, assign multiple labels to each example, reorder candidate documents by relevance on MTEB Multilingual v2 · mean over tasks in 1037 languages including 'Auhelawa, Abau, Achi, Achinese, Achuar-Shiwiar, Adzera, Afrikaans, Agarabi; tasks receive equal weight
Gemini Embedding 00168.4
Gemini Embedding 2—
| Benchmark | Gemini Embedding 001 | Gemini Embedding 2 |
|---|---|---|
| MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) | 68.4 | — |
Specs
| Spec | Gemini Embedding 001 | Gemini Embedding 2 |
|---|---|---|
| Price | $0.15 | $0.20 |
| Context | 2K | 8K |
| Weights | Closed | Closed |
| Released | Jul 2025 | Apr 2026 |
| Dimensions | 3,072 | 3,072 |
| Max input | 2K tokens | 8K tokens |
What breaks when you switch
Gemini Embedding 001 → Gemini Embedding 2
Re-embed your whole corpus›
Vectors from different models can't be mixed.
Gemini Embedding 2 → Gemini Embedding 001
Re-embed your whole corpus›
Vectors from different models can't be mixed.