Choosing an AI model

Admins and owners can choose the model in Settings → AI Model. The choice is stored per organization and applies to every review in it.

Recommended models

The picker lists curated coding-strong models. You can also select any model your gateway advertises, or type a custom identifier. Choosing "use default" reverts to the platform default.

Auditability

The active model is written to the logs on every single review, so you can always determine which model produced a given review. Changing it is recorded in the audit trail with the actor.

Self-hosting

ReviewIQ talks to any OpenAI-compatible gateway, so on a self-hosted deployment you can point it at your own inference endpoint and your code never leaves your infrastructure.