> ## Documentation Index
> Fetch the complete documentation index at: https://docs.prisme.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# RAG Settings
> Configure chunking, embeddings, and retrieval for optimal results
RAG (Retrieval Augmented Generation) settings control how your documents are processed and retrieved. Fine-tuning these settings can significantly improve the quality of agent responses.
## Quick Start: Presets
When creating a knowledge base, choose a preset that matches your needs:
| Preset | Best For | Trade-off |
| ------------ | -------------------------------------- | --------------------------------- |
| **Fast** | Quick setup, general content | Speed over precision |
| **Balanced** | Most use cases | Good balance of speed and quality |
| **Quality** | Complex documents, high accuracy needs | Slower processing, better results |
Each preset configures the parser, chunking strategy, and chunk sizes automatically. You can customize settings after creation if needed.
Start with **Balanced** for most use cases. Switch to **Quality** if you notice retrieval issues with complex documents like PDFs with tables or multi-column layouts.
## Document Parsing
Before chunking, documents are parsed to extract text. The parser affects how well structure is preserved.
| Parser | Identifier | Speed | OCR | Best For |
| ---------------------- | ------------------ | ------- | --- | -------------------------------------- |
| **Tika** | `tika` | Fast | No | Plain text, simple documents |
| **Tika + OCR** | `tika-ocr` | Slow | Yes | Scanned documents, images with text |
| **Unstructured** | `unstructured` | Medium | No | Documents with headings, lists, tables |
| **Unstructured + OCR** | `unstructured-ocr` | Slowest | Yes | Complex scanned layouts, multi-column |
The preset you choose selects an appropriate parser, but you can override it at any time in the knowledge base settings — see [Configuring the Document Parser](#configuring-the-document-parser).
## Understanding RAG
When an agent uses a knowledge base:
1. **Query** - The user's question is converted to an embedding
2. **Search** - Similar document chunks are retrieved
3. **Context** - Retrieved chunks are sent to the AI model
4. **Response** - The model generates an answer using the context
Each step can be configured to optimize for your use case.
## Chunking Settings
Chunking splits documents into smaller pieces for retrieval.
### Chunk Size
How many tokens per chunk (default: 512).
| Size | Pros | Cons |
| ---------------- | ---------------------- | --------------------- |
| **Small (256)** | Precise retrieval | May split context |
| **Medium (512)** | Good balance | Default choice |
| **Large (1024)** | More context per chunk | Less precise matching |
**When to adjust:**
* Decrease for Q\&A with short, specific answers
* Increase for documents where context spans paragraphs
### Chunk Overlap
Tokens shared between consecutive chunks (default: 50).
Overlap ensures that information at chunk boundaries isn't lost. A sentence at the end of one chunk appears at the start of the next.
| Overlap | Pros | Cons |
| ---------------- | ----------------- | --------------------------- |
| **Small (0-25)** | Less redundancy | May lose boundary context |
| **Medium (50)** | Balanced | Default choice |
| **Large (100+)** | Better continuity | More storage, slower search |
### Chunking Strategy
How text is split:
| Strategy | Description |
| ------------- | ------------------------------ |
| **Fixed** | Split at token count (default) |
| **Paragraph** | Split at paragraph boundaries |
| **Sentence** | Split at sentence boundaries |
| **Semantic** | Split by meaning changes |
Start with the default fixed strategy. Switch to semantic chunking if you notice important concepts being split awkwardly.
## Embedding Settings
### Embedding Model
The model that converts text to vectors. Pick it carefully — it cannot be changed in place later.
Consider:
* **Language** - Some models specialize in specific languages
* **Domain** - Specialized models for code, legal, medical, etc.
* **Size** - Larger models are more accurate but slower
The embedding model and its dimensions are **frozen at the moment a knowledge base is created**. The Reindex button reapplies chunking and parsing, but it cannot swap the embedding model — the physical vector index is allocated for the original model's dimensions and cannot be resized. To move to a different embedding model, follow the side-by-side migration in [Changing the embedding model](/products/agent-factory/knowledge-architecture#changing-the-embedding-model-the-a-b-pattern).
### Embedding Dimensions
Higher dimensions capture more nuance but use more storage. Most models have a fixed dimension (e.g., 1536 for OpenAI embeddings). Like the model itself, dimensions are frozen at creation.
## Retrieval Settings
### Top K
How many chunks to retrieve (default: 5).
| K Value | Pros | Cons |
| ----------------- | ------------- | ----------------------------- |
| **Small (3-5)** | Focused, fast | May miss relevant info |
| **Medium (5-10)** | Good coverage | Default choice |
| **Large (10-20)** | Comprehensive | May include irrelevant chunks |
### Similarity Threshold
Minimum similarity score to include a chunk (0-1 scale).
| Threshold | Effect |
| ---------------- | ------------------------------- |
| **Low (0.3)** | More results, lower relevance |
| **Medium (0.5)** | Balanced |
| **High (0.7)** | Fewer results, higher relevance |
Set higher thresholds when precision matters more than recall.
### Reranking
Optional second pass to improve retrieval quality:
1. Initial search retrieves more candidates (e.g., 20)
2. A reranking model scores each candidate
3. Only top results are used
Reranking improves quality but adds latency. Enable for use cases where accuracy is critical.
## Knowledge Base Settings
Access these in the knowledge base's Settings tab.
### Configuring Chunking
1. Open the knowledge base
2. Go to **Settings** > **Processing**
3. Adjust chunk size and overlap
4. Click **Save**
Changing chunking settings only affects new documents. Click **Reindex All** to apply changes to existing documents.
### Configuring Retrieval
1. Open the knowledge base
2. Go to **Settings** > **Retrieval**
3. Adjust Top K, threshold, and reranking
4. Click **Save**
Retrieval settings take effect immediately - no reindexing needed.
### Configuring the Document Parser
The parser can be changed after the knowledge base is created — it is no longer locked to the preset chosen at creation time.
1. Open the knowledge base
2. Go to **Settings** > **Processing**
3. Select the parser — use **Tika + OCR** (`tika-ocr`) for scanned documents or images that contain text
4. Click **Save**, then **Reindex All**
Switching the parser only affects new documents until you click **Reindex All** to reprocess existing ones. OCR parsers (`tika-ocr`, `unstructured-ocr`) extract text from scans and images but are noticeably slower to index.
For files uploaded directly into an agent's chat, the parser lives on the agent's conversation store rather than a regular knowledge base. See [Enabling OCR for chat-uploaded files](/products/agent-factory/knowledge-architecture#enabling-ocr-for-chat-uploaded-files).
## Testing Retrieval
After changing settings, test the impact:
1. Go to the knowledge base
2. Use the **Test Search** feature (if available)
3. Enter a query
4. Review which chunks are retrieved
5. Check if relevant content is included
Or test via an agent:
1. Open an agent that uses this knowledge base
2. Go to Playground
3. Ask questions and observe tool calls
4. Check if the retrieved context is relevant
## Common Scenarios
### FAQ-Style Content
Short questions with specific answers:
* Chunk size: 256-512
* Top K: 3-5
* High similarity threshold (0.6+)
### Long-Form Documents
Research papers, manuals, reports:
* Chunk size: 512-1024
* Overlap: 100+
* Top K: 5-10
### Technical Documentation
Code examples, API references:
* Consider code-aware chunking
* Higher overlap to preserve examples
* Semantic chunking if available
### Mixed Content
Various document types:
* Start with defaults
* Test with representative queries
* Adjust based on results
## Performance Considerations
### Index Size
Smaller chunks with more overlap = larger index:
* More storage used
* Potentially slower search
* Better for accuracy-critical use cases
### Query Latency
Factors affecting search speed:
* Number of documents
* Chunk count
* Top K value
* Reranking enabled
For large knowledge bases, balance quality vs. speed.
### Cost
Consider:
* Embedding API costs for indexing
* Reranking costs if enabled
* Token costs from larger context
## Troubleshooting
### "Agent doesn't find relevant content"
* Decrease similarity threshold
* Increase Top K
* Check if content is actually indexed
* Verify the query matches document language/terminology
### "Retrieved context seems irrelevant"
* Increase similarity threshold
* Enable reranking
* Review chunk boundaries
* Consider different embedding model
### "Important context is split across chunks"
* Increase chunk size
* Increase overlap
* Try semantic chunking
### "Too much irrelevant content in responses"
* Decrease Top K
* Increase similarity threshold
* Enable reranking
* More specific instructions in agent prompt
## Next Steps
See how settings affect real queries in the Playground
Learn about multi-query, hierarchical retrieval, and more