← All benchmarks

InteractiveBench Finding (task pass)

Agents · 2026-10-25

InteractiveBench's Finding category: the 20 Finding tasks, in which the assistant must remember where objects were moved and guide the search for them (for example tools stored in a lidded box inside a cabinet in another room). Same simulator, Gemini 3.1 Pro Human agent and end-state check as the full benchmark; the score is the percentage of these 20 tasks in which every required end state holds.

Top models (higher is better)