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

# Knowledge Config • AI Agent SDK

> KnowledgeConfig: Configuration for RAG and knowledge retrieval.

# KnowledgeConfig

> Defined in the [**Feature Configs**](../modules/feature_configs) module.

<Badge color="blue">AI Agent</Badge>

Configuration for RAG and knowledge retrieval.

Consolidates: knowledge, retrieval\_config, knowledge\_config, rag\_config, embedder\_config

## Properties

<ResponseField name="sources" type="List">
  No description available.
</ResponseField>

<ResponseField name="embedder" type="str">
  No description available.
</ResponseField>

<ResponseField name="embedder_config" type="Optional">
  No description available.
</ResponseField>

<ResponseField name="chunking_strategy" type="Union">
  No description available.
</ResponseField>

<ResponseField name="chunk_size" type="int">
  No description available.
</ResponseField>

<ResponseField name="chunk_overlap" type="int">
  No description available.
</ResponseField>

<ResponseField name="chunker" type="Optional">
  No description available.
</ResponseField>

<ResponseField name="retrieval_k" type="int">
  No description available.
</ResponseField>

<ResponseField name="retrieval_threshold" type="float">
  No description available.
</ResponseField>

<ResponseField name="rerank" type="bool">
  No description available.
</ResponseField>

<ResponseField name="rerank_model" type="Optional">
  No description available.
</ResponseField>

<ResponseField name="auto_retrieve" type="bool">
  No description available.
</ResponseField>

<ResponseField name="vector_store" type="Optional">
  No description available.
</ResponseField>

<ResponseField name="config" type="Optional">
  No description available.
</ResponseField>

<Accordion title="Internal & Generic Methods">
  * **to\_dict**: Convert to dictionary.
</Accordion>

## Usage

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
# Simple enable with sources
    Agent(knowledge=["docs/", "data.pdf"])
    
    # With config
    Agent(knowledge=KnowledgeConfig(
        sources=["docs/"],
        embedder="openai",
        chunking_strategy="semantic",
        retrieval_k=5,
        rerank=True,
    ))
    
    # With chunker config
    Agent(knowledge={
        "sources": ["docs/"],
        "chunker": {
            "type": "semantic",
            "chunk_size": 512
        }
    })
```

## Source

<Card title="View on GitHub" icon="github" href="https://github.com/MervinPraison/PraisonAI/blob/main/src/praisonai-agents/praisonaiagents/config/feature_configs.py#L208">
  `praisonaiagents/config/feature_configs.py` at line 208
</Card>

***

## Related Documentation

<CardGroup cols={2}>
  <Card title="Knowledge Concept" icon="book" href="/docs/concepts/knowledge" />

  <Card title="Knowledge Overview" icon="database" href="/docs/knowledge/overview" />

  <Card title="Knowledge Configuration" icon="gear" href="/docs/configuration/knowledge-config" />

  <Card title="Chat with PDF" icon="file-pdf" href="/docs/knowledge/chat-with-pdf" />

  <Card title="Configuration Overview" icon="gear" href="/docs/configuration/index" />
</CardGroup>