> ## Documentation Index
> Fetch the complete documentation index at: https://docs.noxus.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Advanced Settings

> Configure search methods and retrieval settings

Fine-tune knowledge base behavior with advanced configuration options.

<img src="https://mintcdn.com/spot-16018069/aIe-Q8xSXPJ4vFiE/images/kbs/retrieval.png?fit=max&auto=format&n=aIe-Q8xSXPJ4vFiE&q=85&s=ae444c88e2612f75cf0b9ded1fe62c5e" alt="Retrieval Settings" width="1085" height="660" data-path="images/kbs/retrieval.png" />

## Search Methods

<Tabs>
  <Tab title="Semantic Search">
    **Uses vector embeddings to understand meaning**

    | Configuration        | Options                         | Default   |
    | :------------------- | :------------------------------ | :-------- |
    | Embedding model      | FastEmbed, OpenAI, Multilingual | FastEmbed |
    | Similarity threshold | 0.0 - 1.0                       | 0.7       |
    | Top-K results        | 1 - 50                          | 5         |

    <Info>
      **Best For:** Natural language queries, conceptual similarity, finding related information
    </Info>

    **Example:** "climate initiatives" matches "environmental programs", "sustainability efforts"
  </Tab>

  <Tab title="Keyword Search">
    **Traditional keyword matching with BM25 ranking**

    | Configuration   | Options           | Default        |
    | :-------------- | :---------------- | :------------- |
    | BM25 parameters | k1, b values      | k1=1.2, b=0.75 |
    | Stop words      | Enable/disable    | Enabled        |
    | Stemming        | Language-specific | English        |

    <Info>
      **Best For:** Exact term searches, technical terminology, IDs and codes
    </Info>

    **Example:** "SKU-12345" requires exact match
  </Tab>

  <Tab title="Hybrid Search">
    **Combines semantic + keyword search**

    | Configuration   | Options               | Default  |
    | :-------------- | :-------------------- | :------- |
    | Alpha parameter | 0.0 - 1.0             | 0.5      |
    | Fusion method   | RRF, Relative scoring | RRF      |
    | Reranking       | Enable/disable        | Disabled |

    <Tip>
      **Recommended:** Start with alpha=0.5 for balanced results between keyword and semantic search
    </Tip>

    **Best For:** Most general-purpose queries with balance between precision and recall
  </Tab>

  <Tab title="Reranking">
    **Second-stage ranking for improved precision**

    **How It Works:**

    1. Retrieves larger candidate set (e.g., top 50)
    2. Uses ColBERT model to rerank
    3. Returns top-K most relevant (e.g., top 5)

    <Warning>
      **Trade-off:** Slower but more accurate. Use when precision is critical.
    </Warning>
  </Tab>
</Tabs>

## Retrieval Settings

<AccordionGroup>
  <Accordion title="Top-K (Number of Chunks)" icon="list">
    How many chunks to retrieve:

    **Recommendations:**

    * **5-10 chunks**: Most use cases
    * **3-5 chunks**: Quick answers, cost-sensitive
    * **10-20 chunks**: Complex questions, comprehensive answers

    **Trade-offs:**

    * More chunks = more context but slower and costlier
    * Fewer chunks = faster but may miss information
  </Accordion>

  <Accordion title="Similarity Threshold" icon="gauge">
    Filter out low-relevance chunks:

    **Range:** 0.0 (all results) to 1.0 (exact match only)

    **Recommendations:**

    * **0.5-0.6**: Lenient, more results
    * **0.7**: Balanced, good default
    * **0.8+**: Strict, high precision
  </Accordion>

  <Accordion title="Chunk Size" icon="maximize">
    Size of document segments:

    **Options:**

    * **256 tokens**: Precise, more chunks needed
    * **512 tokens**: Balanced, recommended default
    * **1024 tokens**: More context per chunk

    **Considerations:**

    * Smaller chunks = more precise but need more retrievals
    * Larger chunks = more context but less precise
  </Accordion>

  <Accordion title="Chunk Overlap" icon="layers">
    Overlap between consecutive chunks:

    **Typical:** 50-100 tokens

    **Purpose:**

    * Prevents information loss at boundaries
    * Ensures continuity of context
    * Improves retrieval quality
  </Accordion>
</AccordionGroup>

## Embedding Models

| Model                   | Quality | Speed | Cost   | Best For                                |
| :---------------------- | :------ | :---- | :----- | :-------------------------------------- |
| **FastEmbed** (Default) | ⭐⭐⭐     | ⚡⚡⚡   | 💰     | Most general use cases                  |
| **OpenAI Embeddings**   | ⭐⭐⭐⭐⭐   | ⚡⚡    | 💰💰💰 | Critical applications, maximum accuracy |
| **Multilingual Models** | ⭐⭐⭐⭐    | ⚡⚡    | 💰💰   | International documents, 100+ languages |

<Note>
  FastEmbed provides the best balance of quality, speed, and cost for most use cases. Upgrade to OpenAI for mission-critical applications or multilingual models for international content.
</Note>

## Advanced Features

<AccordionGroup>
  <Accordion title="Folder-Based Search" icon="folder">
    Limit search to specific folders:

    **Benefits:**

    * Faster searches
    * More relevant results
    * Domain-specific retrieval
    * Organized knowledge

    **Use Cases:**

    * Search only "HR Policies" for HR questions
    * Search only "Product Docs" for technical questions
    * Separate public vs internal documents
  </Accordion>

  <Accordion title="Metadata Filtering" icon="filter">
    Filter by custom metadata fields:

    **Examples:**

    * Only documents from "Engineering" department
    * Only documents tagged "2024"
    * Only documents by specific author
    * Only documents of type "Policy"

    **Configuration:**

    ```json theme={null}
    {
      "metadata_filter": {
        "department": "Engineering",
        "year": "2024",
        "doc_type": "Technical Spec"
      }
    }
    ```
  </Accordion>

  <Accordion title="Citation Configuration" icon="quote">
    Control how sources are cited:

    **Options:**

    * Include page numbers
    * Show file names
    * Display chunk IDs
    * Add custom metadata in citations

    **Example Output:**

    ```
    Source: Installation Guide.pdf, Page 3
    Department: Engineering
    Last Updated: 2024-01-15
    ```
  </Accordion>
</AccordionGroup>

<Card title="Managing Through Flows" icon="workflow" href="/platform/knowledge-bases/managing-flows">
  Automate knowledge base operations with flows
</Card>