Gemini CLI
gemini-cliGemini CLI's pass rate on every board it ran on. Scores on different boards don't add up, so this page doesn't add them.
On each board
Terminal-Bench 2.0
Every board here: version not recorded, full, 89 tasks, 5 attempts, openRank 3
3 harnesses59.8%51.5–68.1Gemini CLI minus tongagents 0.8.0: −20.4 points, 95% interval −28.3 to −13.0, over 89 tasks both ran; a clear difference.
3 harnesses59.8%51.5–68.1Gemini CLI minus tongagents 0.8.0: −20.4 points, 95% interval −28.3 to −13.0, over 89 tasks both ran; a clear difference.
47.4%38.7–55.7