[KongchangAI]
BenchmarkTerminal-Bench 4.1

Terminal-Bench 4.1

A subsequent version of Terminal-Bench, used to evaluate the multi-step task execution capabilities of agents in terminal environments.

Core Facts

Timeline (last 90 days)

Nov 13

Gemini 4 Hydrogen has a 57% pass rate on Terminal-Bench 4.1.

Partially Verified75%
Nov 12

Migrating from Gemini 3.8 Flash to Hydrogen, at the default high setting, task completion speed increases 2.5 times, the Terminal-Bench 4.1 score rises from 18.2% to 57%, and fabricated claims decrease by more than 5 times.

Verified90%

Source Articles