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

Multi-turn Tool-use on FTRL

42.41Solve Rate (R)

PACT

20.736426.363231.9937.6168Jun 15, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
42.4134.6936.6
2026.06
37.1730.432.03
2026.06
36.130.631.6
2026.06
33.1231.830.84
2026.06
32.6929.8429.88
2026.06
32.5928.9828.28
2026.06
32.1431.6628.6
2026.06
31.9833.2431.34
2026.06
28.3326.9422.93
2026.06
28.1929.3224.67
2026.06
24.1221.521.43
2026.06
23.3919.6419.72
2026.06
22.7917.9317.99
2026.06
22.7822.0621.06
2026.06
22.6819.1418.5
2026.06
22.3617.8917.66
2026.06
21.6620.7819.72
2026.06
21.5717.3718.13