Tools

Message Info

Model badge

Each assistant message displays the AI model that generated it in the upper-right corner. This helps you track which model was used, especially if you switch providers mid-conversation.

Token count

The upper-left corner of each assistant message shows the output token count — the number of tokens the AI used to generate that specific response. This does not include input tokens (your conversation history), only what the model produced.

Tip

Longer conversations cost more per message because the full conversation history is sent as input each time. The token badge only shows output tokens, so actual API cost per message is higher than what the badge displays.