Open Models Almanac
Methodsevaluationbenchmark

single mention · evaluation

Computer-use closed-loop evaluation

Evaluation in which a model inspects and interacts with an existing application to recreate a functional version.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See benchmark for the methods filed beside it.

Also called Computer use closed loop.

Filed under evaluation › benchmark