NEWVectors or files. Pick a path.Start →
    Models/Reranking/jinaai/jina-reranker-v3
    HFRerankingcc-by-nc-4.0

    jina-reranker-v3

    by jinaai

    Compact Qwen3-based reranker for multilingual text retrieval precision

    1.1Mdl/month
    143likes
    597Mparams
    Identifiers
    Model ID
    jinaai/jina-reranker-v3
    Feature URI
    mixpeek://reranker@v1/jina_reranker_v3

    Overview

    Jina Reranker v3 is a production text-ranking model from Jina AI built on Qwen3-0.6B. It is useful as a second-stage reranker after dense, sparse, or hybrid retrieval, especially when agents need cleaner evidence from transcripts, OCR spans, documents, and other text-heavy payloads.

    On Mixpeek, Jina Reranker v3 fits after a broad first-stage search. Retrieve a larger candidate set with embeddings, BM25, or hybrid search, then rerank the candidates before sending compact evidence to an agent. This is particularly useful when a video or document pipeline has many short evidence spans that look similar in embedding space.

    Architecture

    Transformer cross-encoder text-ranking model based on Qwen3-0.6B. The model scores query and document pairs directly, so inference cost scales with the number of candidate pairs reranked.

    Mixpeek SDK Integration

    import { Mixpeek } from "mixpeek";
    
    const mx = new Mixpeek({ apiKey: "API_KEY" });
    
    // Managed: create a collection over a bucket; Mixpeek runs this model's extractor
    const collection = await mx.collections.create({
      namespace_id: "my-namespace",
      collection_name: "my-collection",
      source: { type: "bucket", bucket_ids: ["bkt_your_bucket"] },
      feature_extractor: {
        feature_extractor_name: "text_embeddings",
        version: "v1",
        parameters: { model_id: "BAAI/bge-m3" },
      },
    });

    Capabilities

    • Second-stage reranking for dense, sparse, and hybrid candidate sets
    • Multilingual text-ranking model based on Qwen3-0.6B
    • Useful for transcript, OCR, caption, and document span precision
    • Compact 596M-parameter footprint compared with larger rerankers
    • Works well as the precision stage before agent context assembly

    Use Cases on Mixpeek

    Rerank transcript spans from video and audio search before an agent answers
    Rerank OCR blocks from screenshots, PDFs, and slide decks
    Improve precision for hybrid search over support tickets, docs, and media captions
    Select the best evidence packets before payload projection and citation generation

    Benchmarks

    DatasetMetricScoreSource
    Hugging FacePipeline tagtext-rankingJina AI, 2026 Model Card
    Hugging FaceBase modelQwen3-0.6BJina AI, 2026 Model Card
    Hugging FaceParameters596MJina AI, 2026 Model Card

    Performance

    Input SizeText query and candidate text pairs
    GPU LatencyScales with reranked candidate count
    GPU ThroughputUse after top-K candidate generation
    GPU MemoryCompact 596M-parameter reranker

    Specification

    FrameworkHF
    Organizationjinaai
    FeatureReranking
    OutputRelevance score per candidate
    Modalities
    RetrieverCross-Modal Reranker
    Parameters597M
    Licensecc-by-nc-4.0
    Downloads/mo1.1M
    Likes143

    Research Paper

    Jina Reranker v3

    arxiv.org

    Build a pipeline with jina-reranker-v3

    Add this model to a processing pipeline alongside other extractors. Combine with retrieval stages for end-to-end search.

    Run it on your own data, free