Evals / image

Text and logos

3 scenarios, 3 run by an agent. Each run shows its result, time and new tokens.

ScenarioCodexwithout deftCodexwith deft, skill installedCodexwith deft, skill as agents.mdClaude Opus 5.5without deftClaude Opus 5.5with deft, skill installedClaude Codewithout deftClaude Codewith deft
Put a caption on a photoAdd the text Sample in the bottom right corner of DSCN0010.jpg and save it as captioned.jpg.pass15s · 6.5kpass20s · 9.8kpass12s · 6.0kpass15s · 10.2kpass15s · 11.0knot runnot run
Add a Chinese caption to a photo在 DSCN0010.jpg 右下角加上“版权所有”四个字,保存为 captioned.jpg。pass20s · 7.5kpass16s · 15.9kpass11s · 5.8kpass16s · 10.1kpass20s · 11.1knot runnot run
Put a logo on a photoPlace dice.png as a small logo in the top left corner of DSCN0010.jpg. Save it as branded.jpg.pass15s · 3.7kpass17s · 8.7kpass16s · 10.5kpass13s · 9.7kpass15s · 12.0knot runnot run