jina-embeddings-v5-text: Task-Targeted Embedding Distillation

Mohammad Kalim Akram, Saba Sturua, Nastia Havriushenko, Quentin Herreros, Michael Günther, Maximilian Werk, Han Xiao · Feb 17, 2026 · Citations: 0

Automatic Metrics General

Open arXiv RSS feed

Abstract

Text embedding models are widely used for semantic similarity tasks, including information retrieval, clustering, and classification. General-purpose models are typically trained with single- or multi-stage processes using contrastive loss functions. We introduce a novel training regimen that combines model distillation techniques with task-specific contrastive loss to produce compact, high-performance embedding models. Our findings suggest that this approach is more effective for training small models than purely contrastive or distillation-based training paradigms alone. Benchmark scores for the resulting models, jina-embeddings-v5-text-small and jina-embeddings-v5-text-nano, exceed or match the state-of-the-art for models of similar size. jina-embeddings-v5-text models additionally support long texts (up to 32k tokens) in many languages, and generate embeddings that remain robust under truncation and binary quantization. Model weights are publicly available, hopefully inspiring further advances in embedding model development.

Human Data Lens

Uses human feedback: No
Feedback types: None
Rater population: Unknown
Unit of annotation: Unknown
Expertise required: General

Evaluation Lens

Evaluation modes: Automatic Metrics
Agentic eval: None
Quality controls: Not reported
Confidence: 0.40
Flags: low_signal, possible_false_positive

Research Summary

Contribution Summary

Text embedding models are widely used for semantic similarity tasks, including information retrieval, clustering, and classification.
General-purpose models are typically trained with single- or multi-stage processes using contrastive loss functions.
We introduce a novel training regimen that combines model distillation techniques with task-specific contrastive loss to produce compact, high-performance embedding models.

Why It Matters For Eval

Benchmark scores for the resulting models, jina-embeddings-v5-text-small and jina-embeddings-v5-text-nano, exceed or match the state-of-the-art for models of similar size.