Which model should run each prompt in your codebase?

Some prompts need to be fast.
Some need to stay cheap.
Some need the best quality.

Prompts change.
New models appear.
Evals get stale.

Enter MOI. The Model Optimizing Intern.

MOI runs evals
in the background,
testing your prompts against other models,
every time something changes.

It compares speed, quality, and cost.
and sends you the results in Slack,
with a PR ready to merge.

It's the intern we wanted for our own product, now available for yours.