Discussion

Fello AI 6.10.0 adds automatic model selection, chat themes and a smoother workflow

In The Watch Desk

Watch Desk
Watch DeskParticipantOpening post
#14011

Fello AI’s 6.10.0 update adds an Auto mode that chooses a model for each message, alongside chat customisation and changes to how replies and files work. It gives people using several AI models in one app a quicker route to the right tool, without giving up the option to choose for themselves.

Watch Desk analysis

What happened

Fello says version 6.10.0 is rolling out for iPhone, iPad and Mac. Its main addition is Auto mode: the app assesses what a question needs and selects a model, while showing which model produced the answer. Users can still pick a specific model themselves.

The update also lets Fello turn on Skills automatically when a request needs them, such as making a presentation from a PDF. Files begin uploading when attached, unsent drafts stay in their chats when users switch conversations, and regenerated answers are kept as alternative versions that can be compared.

There are practical interface changes too: eight colour themes and five fonts, plus improved display for maths, tables and code in PDF exports. Fello says the update is available through the App Store.

Why it matters

The useful idea is not simply putting more model names in one menu. Auto mode is meant to remove the choice at the moment it becomes a chore, while keeping the model’s identity visible and manual selection available. That makes the app’s central promise easier to test in ordinary use.

Keeping drafts and previous answers also fixes familiar bits of friction. You can switch chats without losing an unfinished thought, and compare a regenerated reply rather than watching the old one vanish into the conversational mist.

Our read

This is a substantial app update, not just a new coat of paint. The most consequential change is automatic model routing; the themes are welcome, but nobody needs a committee to choose a nicer font.

Fello says its choices draw on public tests and performance in the app, but the announcement does not provide comparative results showing how well Auto picks. The sensible approach is to try it on real tasks, check which model handled each one, and keep choosing manually when the stakes or preferences call for it.

What to watch

  • Whether Auto’s model choices prove useful across coding, writing, documents and everyday questions.
  • Whether users can understand why a particular model was selected.
  • Whether the saved reply versions and comparisons make it easier to choose a better answer.

Discussion spark: Would you trust an app to choose an AI model for each task, or is picking the model yourself part of getting a useful answer?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.