Embedding / RAG90

Gemini embedding 004

Company Google

Model type Embedding

Shipping Gemini-class embedding endpoint for Vertex AI and Gemini API retrieval stacks.

PricingPaid / API

Accesshttps://ai.google.dev/gemini-api/docs/embeddings

Updated2026-05-06

Best for

GCP-native RAG, multimodal-ish retrieval stacks, and batch embedding.

About this listing

Scores reflect a curated, builder-oriented ranking for comparing models in product workflows—not a live API benchmark. Verify pricing and terms on the provider site before production use.

Similar models

Other models in adjacent categories or with nearby scores.

Embedding / RAG92

text-embedding-4-large

OpenAI

Type Embedding

OpenAI’s latest large embedding model tuned for retrieval, deduping, and RAG backends.

Best forEnterprise RAG, semantic search, and hybrid vector indexes.

Pricing Paid / API

Embedding / RAG88

Cohere Embed v4

Cohere

Type Embedding

Latest multilingual Cohere embed family for retrieval, rerank, and search stacks.

Best forSearch ranking, classification features, and Cohere-first stacks.

Pricing Freemium / API

Embedding / RAG84

BGE M3

BAAI

Type Embedding

Flagship multilingual BGE checkpoints for dense, sparse, and hybrid retrieval setups.

Best forOpen-source RAG, local vector DBs, and academic baselines.

Pricing Open weights