Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Agentic Tool-Use on ACEBench

37.5ACE-E Score

GEAR

21.925.953034.05May 12, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2026.05
37.545.3-
2026.05
36.338.9-
2026.05
33.737.8-
2026.05
32.835.9-
2026.05
32.536.3-
2026.05
32.536.4-
2026.05
31.535.7-
2026.05
31.335.8-
2026.05
31.235.6-
2026.05
30.733.4-
2026.05
30.532.9-
2026.05
26.530.2-
2026.05
26.528.8-
2026.05
24.328.1-
2026.05
23.626.2-
2026.05
22.526.5-
2026.05
--0.75
2026.05
--0.325
2026.05
--0.6
2026.05
--0.633
2026.05
--0.65
2026.05
--0.408
2026.05
--0.391
2026.05
--0.634