TypeSafe AI puts Jev's first bounding-box output behind an API flag

Jev 1.14.0 enters preview with longer-context access, while CEO Diogo Almeida keeps image and bounding-box features behind experimental API flags.

By · Published

Primary source: TypeSafe AI

Why it matters

TypeSafe is opening Jev to image and spatial tasks while withholding a production reliability guarantee. The release tests whether developers can use typed AI decisions beyond simple classification while maintaining the reliability the company promises.

AI-generated illustration of a camera beside a monitor displaying a mountain lake, with a red rectangle around a peak and image thumbnails along the screen.

TypeSafe AI, the company co-founded by Diogo Almeida, released Jev 1.14.0 on October 14th with a bounding-box output and opt-in access to longer context, according to its release post. TypeSafe's model documentation also lists Jev 1.14 as available from October 14th. Jev 1.13.0 already supported image inputs.…

Reader comments

Conversation for this story loads after sign-in.