Skip to main content

​
Match the depth of thinking to the job

The response modes Automatic, Fast and Expert (plus Deep Intelligence) with their descriptions
Every prompt you send Chief runs in a response mode that decides how much reasoning to spend on your answer. Quick questions should come back fast. Hard, multi-step questions deserve deeper thinking. Response modes let you tune that trade-off — or hand the decision to Chief. You pick the mode from the Intelligence chip in the prompt bar, next to where you type. There are three everyday modes:
  • Automatic — Chief switches between Fast and Expert based on the complexity of your prompt. This is the default, and the right choice most of the time.
  • Fast — prioritizes speed over depth. Best for quick lookups, simple rewrites, and short follow-ups where you want an answer immediately.
  • Expert — deeper reasoning, but not as fast. Best for analysis, comparisons, drafting, and anything where you’d rather wait a moment for a stronger answer.
💡 Good to Know: When you leave the mode on Automatic, Chief reads your prompt and routes it to Fast or Expert for you — so you don’t have to think about modes at all unless you want to.

​
How to choose a mode

The Intelligence mode selector chip in the prompt bar, set to Automatic
  1. In the prompt bar, find the mode selector — the Intelligence chip, which shows the current mode (e.g. Automatic) — next to the message box. (If the toolbar is collapsed, click Using: to expand it.)
  2. Click it to open the list of modes.
  3. Select Automatic, Fast, or Expert.
  4. Type your prompt and send it. The mode you picked applies to that prompt.

​
Which mode should I use?

  • Not sure? Leave it on Automatic.
  • Want the answer back as quickly as possible? Use Fast.
  • Want the most thorough single-model answer? Use Expert.
⚠️ Important: Expert mode uses roughly 3x the credits of a standard response because it does deeper reasoning. There’s also a more powerful Deep Intelligence mode for the hardest, multi-agent work.

​
FAQ

Automatic lets Chief decide between Fast and Expert per prompt based on how complex your request looks. Picking Fast or Expert yourself locks that mode in regardless of the prompt, which is useful when you already know whether you want speed or depth.
The mode applies to the prompt you send. If you want a different depth for your next question, open the mode selector again and change it before sending.
Expert prioritizes depth of reasoning, not a fixed wait time. For a simple question it may still answer quickly. If you need the heaviest, multi-agent analysis, use Deep Intelligence mode instead.