​
What is Knowledge?

Knowledge is Relevance AI’s RAG solution that allows you to provide additional context and relevant information to your agents and tools. It serves as the memory and information repository that your AI workforce can access when responding to queries or performing tasks. With Knowledge, your agents can draw upon specific information rather than relying solely on their pre-trained knowledge.

​
Why Use Knowledge?

Adding knowledge to your agents and tools offers several key benefits:
  • Enhanced accuracy - Agents can reference specific information rather than generating responses based only on their training data
  • Reduced hallucinations - By grounding responses in factual information, agents are less likely to generate incorrect or fabricated answers
  • Domain-specific expertise - Easily make your agents experts in particular subjects by providing relevant knowledge bases
  • Up-to-date information - Keep your agents informed with the latest information by updating your knowledge bases

​
Creating Knowledge in Relevance AI

Creating knowledge in Relevance AI is simple and flexible with multiple options to suit your needs:

​
Manually Input Data

Start with an empty table and enter information directly. This approach works well when you have specific information you want to make available to your agents that isn’t contained in existing documents. The manual approach gives you complete control over the structure and content of your knowledge base, allowing you to create precisely tailored information resources.

​
Upload Structured and Unstructured Data

Import files such as CSV, PDF, Excel, JSON, or even audio files. This method is ideal when your information already exists in document form and you want to make it accessible to your agents. Relevance AI automatically processes these documents, extracting the relevant information and making it searchable for your agents. This capability allows you to quickly transform existing documentation into actionable knowledge.

​
Extract Content from Websites

Fetch text and relevant information directly from URLs. This approach is perfect for incorporating information from your company website, documentation sites, or other online resources. Website extraction saves you the time of manually copying information and ensures your knowledge base stays synchronized with your online content.

​
Connect Third-Party Integrations

Sync data from external platforms to create comprehensive knowledge bases that draw from multiple sources. This integration capability allows you to connect your existing tools and data repositories with Relevance AI. By connecting to third-party platforms, you can ensure your agents have access to the most current information across your entire technology stack.

​
Accessing Knowledge

Once you’ve created knowledge bases, you can connect them to your agents and tools. This connection allows your AI workforce to:
  1. Search through relevant information when responding to queries
  2. Reference specific facts and details
  3. Provide responses grounded in your organization’s specific information
  4. Deliver more accurate and contextually appropriate answers

​
Advanced Knowledge Features

Relevance AI’s knowledge capabilities include:
  • Semantic search - Find information based on meaning, not just keywords
  • Knowledge enrichment - Enhance your knowledge bases with additional context using tools
  • Advanced retrieval options - Configure how your agents search and prioritize information
  • Versioning and updates - Keep your knowledge bases current with easy updating

​
Getting Started with Knowledge

To begin using Knowledge in Relevance AI:
  1. Navigate to the Knowledge section in your Relevance AI dashboard
  2. Choose your preferred method for creating a knowledge base
  3. Upload or input your information
  4. Connect your knowledge base to relevant agents and tools
  5. Test your agents to ensure they’re effectively utilizing the knowledge
By following these steps, you’ll create a more capable AI workforce that can leverage specific information to deliver better results. For more information on working with Relevance AI’s key concepts:
  • Tools: Build integrations, LLM prompt chains, or other step-by-step automations
  • Agents: Create reasoning bots powered by LLMs that plan and complete tasks
  • API keys: Enter your own API key(s) for the many supported vendors