NEWVectors or files. Pick a path.Start →
    Advanced Search

    Search with Neural Reranking

    Semantic search followed by cross-encoder reranking and formatted output.

    No credit card to start · Build from $25/mo

    Pipeline Stages

    1
    initial_search
    filter
    feature_search
    View parameters
    {
      "searches": [
        {
          "feature_uri": "mixpeek://text_extractor@v1/multilingual_e5_large_instruct_v1",
          "query": {
            "input_mode": "text",
            "value": "{{INPUT.query}}"
          },
          "top_k": 100
        }
      ],
      "final_top_k": 50
    }
    2
    rerank
    sort
    rerank
    View parameters
    {
      "inference_name": "baai_bge_reranker_v2_m3",
      "query": "{{INPUT.query}}",
      "document_field": "content",
      "top_k": 10,
      "batch_size": 32
    }
    3
    format_output
    apply
    json_transform
    View parameters
    {
      "template": "{\"id\": \"{{DOC.document_id}}\", \"content\": \"{{DOC.content}}\", \"score\": {{DOC.scores.rerank}}}",
      "fail_on_error": false
    }

    Input Schema

    query
    text
    required

    Search query text

    e.g. how to train neural networks

    Use Cases

    • Question answering
    • Conversational search
    • High-precision retrieval

    Requirements

    • Collection must have text embeddings indexed
    • Reranker inference service must be available

    Tags

    reranking
    cross-encoder
    advanced

    Ready to use this template?

    Clone it into your namespace and start querying in minutes.