Skip to main content

Model Selection

Model Selection

Each AI provider offers multiple models with different capability and cost profiles. You can select the model for each provider independently from the Provider tab.

OpenAI Models

Model Speed Cost Best for
gpt-4o Fast Medium Recommended default
gpt-4-turbo Medium Higher More thorough responses
gpt-3.5-turbo Fastest Lowest High-volume testing only

Anthropic Models

Model Speed Cost Best for
claude-sonnet-4-5 Fast Medium Recommended default
claude-haiku-4-5 Fastest Lowest High-volume or cost-sensitive
claude-opus-4-5 Slower Highest Maximum quality responses

How to Change the Model

  1. Go to Admin → Provider tab
  2. Enter the model name in the OpenAI Model or Anthropic Model field
  3. Click Save Provider Settings

The model is applied to all new sessions immediately.

Which Model is Active?

The active model is whichever provider is currently selected. If Anthropic is selected, the Anthropic model field determines which model runs. The OpenAI model field is saved but inactive until you switch providers.

Cost Considerations

Model costs vary significantly. For a live client-facing deployment, claude-sonnet-4-5 (Anthropic) or gpt-4o (OpenAI) provide the best balance of quality and cost. Monitor your API usage dashboard on your provider's portal to track spend.