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

Visual Tool-Use on HRBench 8K

73.7Accuracy

ARM-Thinker-7B

57.68461.8426670.158Nov 24, 2025Nov 25, 2025Nov 27, 2025Nov 29, 2025Nov 30, 2025Dec 2, 2025Dec 4, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
73.7-
2025.12
73.3-
2025.11
71.313.3
2025.11
71.21.2
2025.12
70.4-
2025.12
69.9-
2025.12
69.5-
2025.11
69.16.7
2025.11
68.57.6
2025.12
68.4-
2025.12
66.9-
2025.12
64.6-
2025.12
58.3-