Bright runs on two model tiers, and the composer’s model selector chooses between them. Standard is the default, described in the Model menu as “fast, everyday”. Genius runs a more capable frontier model; the menu calls it “deep reasoning, slower”, and it costs more to run. Design your workflow around that tier behavior (default versus frontier, faster versus slower, cheaper versus costlier), because the specific models behind each tier change over time. Open Bright → The composer model selector with Genius active

​
The two tiers

  • Standard is the default tier, used whenever Genius is off.
  • Genius runs a more capable frontier model. It is the slower, deep-reasoning option, and the higher-cost one. The composer flags it while it is active.

​
When to turn Genius on

  • Consequential or hard-to-reverse decisions, where a wrong answer is expensive to unwind.
  • Work with many interacting constraints that a weaker model tends to flatten.
  • Tasks where you cannot easily check the answer yourself.

​
When to leave it off

  • Lookups, navigation, status questions, and other bounded retrieval.
  • Mechanical edits and formatting.
  • Anything you will verify immediately anyway.

​
Tips

  • Pair Genius with the Advisor specialist. It exists to weigh plans and tradeoffs, the same situations where the stronger tier earns its cost.

​
Good to know

  • Standard is the default. Genius is on only when you switch the composer’s model selector to it.
  • The composer flags Genius while it is active. Expect slower replies from it, and a higher cost.
  • The specific models behind each tier change over time, so this page names tiers, not models.

​
Next step

See the build loop these tiers run inside in Loop Engineering.