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

Multimodal Benchmarking on MMMU

78.1Accuracy

Qwen3-VL-32B-Thinking

23.537.67551.8566.025Jul 23, 2024Oct 22, 2024Jan 21, 2025Apr 23, 2025Jul 23, 2025Oct 22, 2025Jan 22, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
78.1
2026.01
74.1
2026.01
68.11
2026.01
62.11
2026.01
60.67
2026.01
59.22
2024.07
37.2
2024.07
37
2024.07
36.4
2024.07
35.8
2024.07
35.6
2024.07
30.6
2024.07
29.7
2024.07
28.8
2024.07
25.6