Jina Embeddings: A Novel Set of High-Performance Sentence Embedding Models

By Michael Gunther, Louis Milliken, Jonathan Geuter, Georgios Mastrapas, Bo Wang, Han Xiao. 2023

Subfields: Information Retrieval & Search, Natural Language Processing, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Retrieval & Search

Jiaxin Zhang

Citations in the corpus

CoRNStack: High-Quality Contrastive Data for Better Code Retrieval and Reranking (2024, unknown): “…models, we consider the Small, Base, and Large variants of CodeSage (Zhang et al., 2024), along with CodeT5+ (Wang et al., 2023b) and Jina-Code-v2 (G¨unther et al., 2023), which are the current state-of-the-art code embedding models on text-to-code retrieval benchmarks. the GitHub Code dataset…”
Beyond Chunk-Then-Embed: A Comprehensive Taxonomy and Evaluation of Document Chunking Strategies for Information Retrieval (2026, unknown): “We evaluate the following models: • jina-embeddings-v2-small-en (Jina-v2) [G¨unther et al.(2023)]: A 33M parameter model optimized for efficiency while maintaining competitive effectiveness.”
EmbeddingRWKV: State-Centric Retrieval with Reusable States (2026, unknown): “First , most state-of-the-art embedding models and rerankers are predominantly built on Transformer architectures (Zhang et al., 2025; Günther et al., 2023).”
SweRank: Software Issue Localization with Code Ranking (2025, unknown): “To mitigate this, we employ filtering and mining techniques following recent work (Wang et al., 2022; Günther et al., 2023; Suresh et al., 2024).”
Do You Know What You Are Talking About? Characterizing Query-Knowledge Relevance For Reliable Retrieval Augmented Generation (2024, background): “…BM25 (Robertson et al., 2009), BGE (Xiao et al., 2023), Contriever (Izacard et al., 2022), MedCPT (Jin et al., 2023), Nomic (Nussbaum et al., 2024), and Jina (Günther et al., 2023), in terms of their ability to distinguish IK and OoK queries and their ability to retrieve relevant documents.”
LocAgent: Graph-Guided LLM Agents for Code Localization (2025, unknown): “Dense retrieval methods require maintaining and continuously updating vector representations of the entire codebase (Wang et al., 2023b; Günther et al., 2023), creating engineering challenges for large, evolving repositories where code changes frequently.”