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

    RAG with Summarization

    Search, rerank, and summarize results into a single answer. Full RAG pipeline for question answering.

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

    Pipeline Stages

    1
    semantic_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": 20
    }
    2
    rerank
    sort
    rerank
    View parameters
    {
      "inference_name": "baai_bge_reranker_v2_m3",
      "query": "{{INPUT.query}}",
      "document_field": "content",
      "top_k": 5
    }
    3
    summarize
    reduce
    summarize
    View parameters
    {
      "provider": "google",
      "model_name": "gemini-2.5-flash-lite",
      "prompt": "Based on the following documents, answer this question: {{INPUT.query}}\n\n{{DOCUMENTS}}\n\nProvide a comprehensive answer:",
      "output_field": "answer",
      "include_sources": true
    }

    Input Schema

    query
    text
    required

    Question to answer

    e.g. What is machine learning?

    Use Cases

    • Question answering systems
    • RAG chatbots
    • Research assistant
    • Knowledge base Q&A

    Requirements

    • Collection must have text embeddings indexed
    • LLM and reranker services must be available

    Tags

    rag
    summarization
    qa
    llm

    Ready to use this template?

    Clone it into your namespace and start querying in minutes.