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

Multimodal Understanding on SEEDBench2 Plus

76.86Accuracy

MIRROR (ours)

6.285624.607842.9361.2522Aug 5, 2025Sep 7, 2025Oct 10, 2025Nov 13, 2025Dec 16, 2025Jan 18, 2026Feb 21, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
76.86---
2026.01
71.19---
2026.01
71.15---
2025.08
71.15---
2026.02
70.88---
2025.08
70.62---
2026.01
70.44---
2026.01
70.44---
2026.02
70.36---
2026.02
70.31---
2026.01
70.0568.460.9783.18
2026.01
69.83---
69.7867.5361.0983.18
2026.01
69.7867.960.7283.18
2025.08
69.74---
2026.01
69.6567.6560.5983.18
2026.02
69.61---
2026.02
69.61---
69.5267.4160.183.64
2025.08
69.43---
2026.01
69.39---
2026.01
69.35---
2026.02
69.3---
2026.01
69.26---
2025.08
69.26---
2026.02
68.99---
2026.01
68.99---
2026.02
68.81---
2026.02
67.9---
2025.08
67.68---
2025.08
67.28---
2026.02
67.19---
2025.08
66.84---
2025.08
66.8---
2025.08
65.74---
2026.02
64.95---
2025.08
64.82---
2025.08
64.82---
2025.08
63.99---
2025.08
63.46---
2026.02
62.32---
2025.08
61.79---
2025.08
58.15---
2025.08
57.27---
2025.08
56.92---
2025.08
56.79---
2025.08
54.98---
2025.08
53.8---
2025.08
53.67---
2025.08
53.32---
2025.08
51.78---
2025.08
51.25---
2025.08
51.25---
2025.08
50.94---
2025.08
48.4---
2026.02
41.15---
2026.02
40.84---
2026.02
40.49---
2026.02
39.75---
2026.02
39.31---
2026.02
38.65---
2026.02
38.21---
2026.02
37.9---
2026.02
35.4---
2026.02
33.38---
2026.02
32.02---
2026.02
31.8---
2026.02
30.87---
2026.02
30.61---
2026.02
28.55---
2026.02
26.39---
2025.08
17.83---
2025.08
9.18---
2025.08
9---