Skip to main content
- Model Selection:
- Available Providers: Agents can be powered by leading models from OpenAI, Gemini, or Anthropic.
- Selection Criteria: Choose your model based on the specific trade-off between reasoning depth (e.g., for complex analysis) and speed/cost (e.g., for simple classification).
- Behavioral Parameters:
- Reasoning Steps: Defines the maximum number of logical sub-tasks the Agent is allowed to perform to reach a conclusion. Higher limits allow for deeper problem-solving but increase latency.
- Temperature: Controls the creativity and variability of the output.
- Low Temperature (e.g., 0.0 - 0.2): Deterministic and precise. Best for data extraction, coding, and classification.
- High Temperature (e.g., 0.7+): Creative and varied. Best for copywriting, brainstorming, or conversational engagement.