Gemini Embedding: Generalizable Embeddings from Gemini

TL;DR

Gemini Embedding is a new multilingual embedding model leveraging Google's Gemini LLM, producing highly generalizable text representations that outperform previous models across diverse languages and tasks.

Contribution

This paper introduces Gemini Embedding, a state-of-the-art multilingual embedding model based on Google's Gemini LLM, demonstrating superior performance on a comprehensive benchmark.

Findings

Outperforms prior models on MMTEB benchmark

Achieves state-of-the-art results in multilingual and code tasks

Demonstrates broad applicability across various downstream tasks

Abstract

In this report, we introduce Gemini Embedding, a state-of-the-art embedding model leveraging the power of Gemini, Google's most capable large language model. Capitalizing on Gemini's inherent multilingual and code understanding capabilities, Gemini Embedding produces highly generalizable embeddings for text spanning numerous languages and textual modalities. The representations generated by Gemini Embedding can be precomputed and applied to a variety of downstream tasks including classification, similarity, clustering, ranking, and retrieval. Evaluated on the Massive Multilingual Text Embedding Benchmark (MMTEB), which includes over one hundred tasks across 250+ languages, Gemini Embedding substantially outperforms prior state-of-the-art models, demonstrating considerable improvements in embedding quality. Achieving state-of-the-art performance across MMTEB's multilingual, English, and code benchmarks, our unified model demonstrates strong capabilities across a broad selection of tasks and surpasses specialized domain-specific models.