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

General MLLM Evaluation on Average V* HR-Bench POPE

84.26Accuracy

SpecEyes

66.153670.854375.55580.2557Mar 24, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.03
84.261.73
2026.03
83.991.42
2026.03
83.780.48
2026.03
82.341.82
2026.03
82.311.8
2026.03
82.291
2026.03
82.091.49
2026.03
81.951.63
2026.03
81.391
2026.03
80.532.31
2026.03
80.531.7
2026.03
78.934.13
2026.03
66.850.43