News
Models
Products
keyboard_arrow_down
DeepSearch
Search, read and reason until best answer found.
Reader
Convert any URL to Markdown for better grounding LLMs.
Embeddings
World-class multimodal multilingual embeddings.
Reranker
World-class reranker for maximizing search relevancy.
More
keyboard_arrow_down
Classifier
Zero-shot and few-shot classification for image and text.
Segmenter
Cut long text into chunks and do tokenization.

API Docs
Auto codegen for your copilot IDE or LLM
open_in_new


Company
keyboard_arrow_down
About us
Contact sales
Intern program
Join us
open_in_new
Download logo
open_in_new
Terms & Conditions


Log in
login
copyright

jina-embeddings-v3

Frontier multilingual embedding model with SOTA performance
Release Postarrow_forward
License
copyright
CC-BY-NC-4.0
Release Date
calendar_month
2024-09-18
Input
abc
Text
arrow_forward
Output
more_horiz
Vector
Model Details
Parameters: 570M
Input Token Length: 8K
Output Dimension: 1024
Language Support
🌍 Multilingual support
Related Models
link
jina-embeddings-v2-base-en
link
jina-embeddings-v2-base-zh
link
jina-embeddings-v2-base-de
link
jina-embeddings-v2-base-es
link
jina-embeddings-v2-base-code
Tags
text-embedding
multilingual
high-dimension
long-context
production
base-model
matryoshka
lora-adapters
Available via
Jina APICommercial LicenseAWS SageMakerMicrosoft AzureGoogle CloudHugging Face
I/O graph
Choose models to compare
Publications (3)
ACL 2025
December 17, 2024
AIR-Bench: Automated Heterogeneous Information Retrieval Benchmark
ECIR 2025
September 18, 2024
jina-embeddings-v3: Multilingual Embeddings With Task LoRA
arXiv
September 07, 2024
Late Chunking: Contextual Chunk Embeddings Using Long-Context Embedding Models

Overview

Jina Embeddings v3 is a groundbreaking multilingual text embedding model that transforms how organizations handle text understanding and retrieval across languages. At its core, it solves the critical challenge of maintaining high performance across multiple languages and tasks while keeping computational requirements manageable. The model particularly shines in production environments where efficiency matters - it achieves state-of-the-art performance with just 570M parameters, making it accessible for teams that can't afford the computational overhead of larger models. Organizations needing to build scalable, multilingual search systems or analyze content across language barriers will find this model especially valuable.

Methods

The model's architecture represents a significant innovation in embedding technology, built on a foundation of jina-XLM-RoBERTa with 24 layers and enhanced with task-specific Low-Rank Adaptation (LoRA) adapters. LoRA adapters are specialized neural network components that optimize the model for different tasks like retrieval, classification, or clustering without increasing the parameter count significantly - they add less than 3% to the total parameters. The model incorporates Matryoshka Representation Learning (MRL), allowing embeddings to be flexibly reduced from 1024 to as low as 32 dimensions while preserving performance. Training involved a three-stage process: initial pre-training on multilingual text from 89 languages, fine-tuning on paired texts for embedding quality, and specialized adapter training for task optimization. The model supports context lengths up to 8,192 tokens through Rotary Position Embeddings (RoPE), with an innovative base frequency adjustment technique that improves performance on both short and long texts.

Performance

The model demonstrates exceptional efficiency-to-performance ratio in real-world testing, outperforming both open-source alternatives and proprietary solutions from OpenAI and Cohere on English tasks while excelling in multilingual scenarios. Most surprisingly, it achieves better results than e5-mistral-7b-instruct, which has 12 times more parameters, highlighting its remarkable efficiency. In MTEB benchmark evaluations, it achieves an average score of 65.52 across all tasks, with particularly strong performance in Classification Accuracy (82.58) and Sentence Similarity (85.80). The model maintains consistent performance across languages, scoring 64.44 on multilingual tasks. When using MRL for dimension reduction, it retains strong performance even at lower dimensions - for example, maintaining 92% of its retrieval performance at 64 dimensions compared to the full 1024 dimensions.

Best Practice

To effectively deploy Jina Embeddings v3, teams should consider their specific use case to select the appropriate task adapter: retrieval.query and retrieval.passage for search applications, separation for clustering tasks, classification for categorization, and text-matching for semantic similarity. The model requires CUDA-capable hardware for optimal performance, though its efficient architecture means it needs significantly less GPU memory than larger alternatives. For production deployment, AWS SageMaker integration provides a streamlined path to scalability. The model excels in multilingual applications but may require additional evaluation for low-resource languages. While it supports long documents up to 8,192 tokens, optimal performance is achieved with the late chunking feature for very long texts. Teams should avoid using the model for tasks requiring real-time generation or complex reasoning - it's designed for embedding and retrieval, not text generation or direct question answering.
Blogs that mention this model
September 18, 2024 • 10 minutes read
Jina Embeddings v3: A Frontier Multilingual Embedding Model
jina-embeddings-v3 is a frontier multilingual text embedding model with 570M parameters and 8192 token-length, outperforming the latest proprietary embeddings from OpenAI and Cohere on MTEB.
Jina AI
Dynamic image showing the characters "V3" formed by bright green dots varying in size on a black background.
May 07, 2025 • 9 minutes read
Model Soup’s Recipe for Embeddings
Boost robustness and performance with model soups: averaging weights. No extra cost, better results.
Bo Wang
Scott Martens
Still life drawing of a purple bowl filled with apples and oranges on a white table. The scene features rich colors against a
April 16, 2025 • 10 minutes read
On the Size Bias of Text Embeddings and Its Impact in Search
Size bias refers to how the length of text inputs affects similarity, regardless of semantic relevance. It explains why search systems sometimes return long, barely-relevant documents instead of shorter, more precise matches to your query.
Scott Martens
Black background with a simple white ruler marked in centimeters, emphasizing a minimalist design.
April 08, 2025 • 21 minutes read
jina-reranker-m0: Multilingual Multimodal Document Reranker
Introducing jina-reranker-m0, our new multilingual multimodal reranker for retrieving visual documents, with SOTA performance on multilingual long documents and code searching tasks.
Jina AI
Modern dot matrix text display on a dark blue background, conveying a digital feel.
March 12, 2025 • 11 minutes read
Snippet Selection and URL Ranking in DeepSearch/DeepResearch
Nailing these two details takes your DeepSearch from mid to GOAT: selecting the best snippets from lengthy webpages and ranking URLs before crawling.
Han Xiao
Logo with words "THINK SEARCH THINK" in black dot and arrow patterns on an orange background, accompanied by horizontal lines
Offices
location_on
Sunnyvale, CA
710 Lakeway Dr, Ste 200, Sunnyvale, CA 94085, USA
location_on
Berlin, Germany (HQ)
PrinzessinnenstraĂźe 19-20, 10969 Berlin, Germany
location_on
Beijing, China
Level 5, Building 6, No.48 Haidian West St. Beijing, China
location_on
Shenzhen, China
402 Floor 4, Fu'an Technology Building, Shenzhen, China
Search Foundation
DeepSearch
Reader
Embeddings
Reranker
Classifier
Segmenter
API Documentation
Get Jina API key
Rate Limit
API Status
Company
About us
Contact sales
Newsroom
Intern program
Join us
open_in_new
Download logo
open_in_new
Terms
Security
Terms & Conditions
Privacy
Manage Cookies
email
Jina AI © 2020-2025.