Chat with your RAG chatbot

💬 Chat with Your Assistant

Once your data is uploaded and your system prompt is set, you can interact with your assistant directly in the Chat tab. This is where you test, explore, and validate how your assistant responds to real questions using your own data.

🧑‍💻 Ask Questions Freely

Type any question into the message bar to start chatting with your assistant. It will use both:

• Your uploaded content (via RAG – Retrieval Augmented Generation)
• Your system prompt instructions (personality, tone, etc.)

You can ask about features, documentation, processes, or anything related to your content.

🗂️ Create and Manage Conversations

Use the Conversation dropdown to:

• Create new conversations
• Switch between previous ones
• Track different test cases or user scenarios

This helps you simulate real interactions and debug any issues with data or prompt design.

🧠 Choose the LLM Model

You can select which Large Language Model (LLM) powers your assistant. The entire list is available here: https://app.edenai.run/models?technology=llm-chat

This gives you flexibility to balance performance, speed, and cost.

⚙️ Customize Model Parameters

Fine-tune how the assistant responds by adjusting the following settings:

• Max tokens: Limit the length of the assistant's response.
• Temperature: Control creativity — lower values make responses more predictable, higher values make them more diverse.
• K: Choose how many top relevant chunks to retrieve from your database when answering.

These settings help you tailor the assistant’s behavior for your specific use case (support, education, technical help, etc.).

Whether you're prototyping or going live, the Chat tab is your playground for building and refining an AI assistant that truly understands your users.