Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Mathematical Reasoning on MathVerse 1.0 (test)
Loading...
73.01
Score
Vanilla (Full Tokens)
16.2988
31.0219
45.745
60.4681
Apr 14, 2026
Score
Updated 3d ago
Evaluation Results
Method
Method
Links
Score
Vanilla (Full Tokens)
Model=Qwen3-VL-4B-Thin...
2026.04
73.01
Vanilla (Full Tokens)
Model=Qwen3-VL-8B, Mod...
2026.04
62.81
DSTP
Model=Qwen3-VL-4B-Thin...
2026.04
58.49
DSTP
Model=Qwen3-VL-8B, Mod...
2026.04
51.16
Vanilla (Full Tokens)
Model=InternVL3.5-14B,...
2026.04
47.89
Vanilla (Full Tokens)
Model=InternVL3.5-8B-T...
2026.04
47.46
DSTP
Model=InternVL3.5-14B,...
2026.04
43.95
FastV (33.3%)
Model=Qwen3-VL-4B-Thin...
2026.04
40.28
FastV (33.3%)
Model=Qwen3-VL-8B, Mod...
2026.04
37.64
DSTP
Model=InternVL3.5-8B-T...
2026.04
37.25
FastV (33.3%)
Model=InternVL3.5-14B,...
2026.04
31.86
FastV (33.3%)
Model=InternVL3.5-8B-T...
2026.04
18.48
Feedback
Search any
task
Search any
task