Embedding Models/Voyage 3 Large

Voyage 3 Large by Voyage AI — 2048D

State-of-the-art general-purpose and multilingual embedding. Outperforms OpenAI-v3-large by 9.74% across 100 datasets. Supports 32K context.

At a glance

Modalities

Dimensions

2K

Max tokens

32K

Parameters

Price / 1M tokens

Type

Dense

Matryoshka dimensions

25651210242048

Output types

Single VectorMulti Vector

Language support

🌍 Multilingual support

Details

Release date 2025-01-01
License Proprietary
Model ID voyage-3-large
Provider Voyage AI

Tags

text-embeddinglong-contextmatryoshkaquantization-aware

What You Need to Know About Voyage 3 Large

Complete Specifications for Voyage 3 Large by Voyage AI

Get detailed specifications for Voyage 3 Large, including output dimensionality of 2048 dimensions, maximum input token length, supported input modalities, and pricing per million tokens. This embedding model is designed for semantic search, text classification, clustering, and retrieval-augmented generation applications where understanding the relationship between texts is essential.

Pricing and Cost Efficiency for Voyage 3 Large

Review the pricing structure for Voyage 3 Large and compare it against other embedding models from Voyage AI and competitors. Understanding embedding model costs is essential when scaling vector search and RAG applications to millions of documents. We provide transparent pricing to help you budget effectively.

Use Cases and Applications for Voyage 3 Large

Explore the ideal use cases for Voyage 3 Large. Whether you are building a semantic search engine, recommendation system, document classification pipeline, or multilingual retrieval system, understanding this model's capabilities and dimensionality will help you choose the right embedding strategy for your project.