Adapt now decides how much intelligence each task needs. Simple, well-defined work runs Fast. Hard, open-ended problems run Deep, with Adapt's most capable reasoning. You don't pick a mode or a model. Adapt reads the work in front of it and matches its intelligence to the task.
How it works
Every request makes different demands. A quick lookup, a long synthesis, a tricky debugging problem, and a costly-to-get-wrong decision each call for a different level of intelligence. Adaptive Intelligence sizes that automatically and shifts as the work changes character, even within a single task.
A quick lookup runs Fast:
A hard, open-ended reasoning problem goes Deep:
This happens behind the scenes on every request, including the work sub-agents do in parallel. There's nothing to configure.
Why it matters
- Better answers on the hard stuff. When a problem is genuinely difficult, Adapt reaches for its deepest reasoning instead of rushing it.
- Fast where speed wins. Routine work stays quick and snappy. You're not waiting on heavy reasoning for a simple lookup.
- Always state-of-the-art. Under the hood, Adapt draws on best-in-class frontier models from Anthropic, Google, and OpenAI, and folds in new ones as they arrive. You get the best of the frontier without ever picking a model.
- No knobs to turn. You ask the same way you always have. Adapt handles the rest.
Availability
Adaptive Intelligence is live for every workspace. No setup required. It applies to all chats, scheduled tasks, apps, and sub-agents automatically.
