Admin

Admin

Draft instructions

Runs when you ask

Reads what you type in the composer so Draft does the right thing with it.

You see the result of this in Inbox

AI connections

Which model does this

Draft understanding

Reads what you typed before Draft runs: finished email, rough text, a partial draft, or an instruction to NephAI. Classifies only; the composer engine writes the email.

Customized

Every model that can actually run this is listed. The one at the top of each group is usually the safe choice.

One of these models always reasons, so Off is unavailable.

Available choices work with both the primary and fallback models.

Thinking harder gives better answers but takes longer and costs more.

900 tokens · about 675 words of room

Reasoning shares this ceiling on some models — Extended gives the answer room to finish.

Longer answers cost more. “Normal” suits almost everything; only raise it if answers are being cut short.

Ordered fallbacks
1.
2.
These choices apply to the next request handled by this service.

How you have told it to approach this

Using NephAI's own wording
Inbox
01
Draft understandingused by Draft understanding

Reads what you typed before Draft runs: finished email, rough text, a partial draft, or an instruction to NephAI.

Default

How much this has been used

Last 28 days

Times it ran

1

Text handled

3,110

What it cost

$0.0073

See this alongside everything else

Other settings that affect this

Writing