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

Audio Reasoning on MMAR

83.7Average Accuracy

Gemini-3.1-Pro

34.92447.58760.2572.913Oct 3, 2025Nov 12, 2025Dec 22, 2025Jan 31, 2026Mar 12, 2026Apr 21, 2026Jun 1, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.06
83.7-------
2026.06
81.73-------
2025.10
7773.3363.5982.9963.6483.0381.7179.17
2025.10
76.472.7361.6584.3572.7381.1978.0583.33
2026.04
71-------
70.7-------
2026.04
69.5-------
2026.06
69.15-------
2026.04
69.1-------
2026.06
68.75-------
68.5-------
2026.06
66.53-------
2026.04
66.4-------
2026.06
66.4-------
2025.10
65.961.2157.2869.0554.5569.2778.0566.67
2025.10
65.661.2150.9772.1181.8272.4865.8570.83
2026.06
64.42-------
2026.06
63.91-------
2025.10
63.553.9450.9770.4163.6472.4862.275
2025.10
63.467.351.564.345.570.264.670.8
2026.06
62.54-------
61.8-------
2026.06
61.7-------
2026.06
60.82-------
2026.04
60.1-------
2026.06
59.78-------
2026.04
59.7-------
2026.04
59.1-------
2026.06
57.96-------
2026.04
57.3-------
2025.10
56.758.7940.7859.8654.5561.9367.0758.33
2026.06
56.7-------
2026.04
56.5-------
2026.06
55.75-------
2026.04
53.7-------
2026.04
51.2-------
2026.06
51.01-------
2025.10
36.843.6433.532.9945.4542.6631.7125