Ai

📸 Screenshots

Here are visual examples of this section:

Ai - Worker Configuration Interface

The AI worker generates text using various AI models from providers like OpenAI, Anthropic, Google, and Groq. It processes prompts, input text, documents, and chat history to produce intelligent responses. This worker supports multiple model providers and allows for contextual conversations with document-based knowledge.

2. Configuration Parameters

temperature: Controls the randomness of the AI output (0 for deterministic, higher values for more creative responses)
model: Specifies the AI model to use in format "provider/model-id" (e.g., "openai/gpt-4", "anthropic/claude-3")

3. Input/Output Handles

prompt: Input handle - accepts the system prompt that defines the AI's behavior and instructions
input: Input handle - accepts the user input or question to be processed
documents: Input handle - accepts vector documents to provide context for the AI response
history: Input handle - accepts chat history to maintain conversation context
answer: Output handle - returns the generated AI response text

4. Usage Examples with Code

// Configure AI worker with OpenAI GPT-4
const aiWorker = {
  parameters: {
    temperature: 0.7,
    model: "openai/gpt-4"
  },
  fields: {
    prompt: { value: "You are a helpful assistant that answers questions based on provided context." },
    input: { value: "What are the key benefits of renewable energy?" },
    documents: { value: relevantDocs },
    history: { value: previousChatHistory }
  }
}

5. Integration Examples

The AI worker serves as a central component in conversational workflows, often following document retrieval workers to provide context-aware responses. It integrates seamlessly with chat interfaces and can be chained with other workers for complex AI-powered automation.

6. Best Practices

Set temperature to 0 for consistent, deterministic outputs in production workflows
Use clear, specific system prompts to guide AI behavior and output format
Provide relevant documents as context to improve response accuracy and relevance
Include chat history for multi-turn conversations to maintain context continuity

7. Troubleshooting Tips

Ensure API keys are properly configured for your chosen model provider
Verify the model format follows "provider/model-id" pattern exactly
Check that the selected model is available and supported by the provider
Monitor token limits when using large documents or extensive chat history

Ai