Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-modal Question Answering on MMMU Pros
Loading...
75
Accuracy
Qwen3.5-27B
34.8768
45.2934
55.71
66.1266
Oct 27, 2025
Nov 23, 2025
Dec 20, 2025
Jan 17, 2026
Feb 13, 2026
Mar 12, 2026
Apr 9, 2026
Accuracy
Updated 6d ago
Evaluation Results
Method
Method
Links
Accuracy
Qwen3.5-27B
Mode=REASONING, Archit...
2026.04
75
Qwen3-VL-235B-A22B
Mode=Thinking, Archite...
2026.04
69.3
EXAONE 4.5 33B
Mode=REASONING, Archit...
2026.04
68.6
Qwen3-VL-32B
Mode=Thinking, Archite...
2026.04
68.1
GPT-5 mini
Mode=REASONING: HIGH,...
2026.04
67.3
MergeMix
baseline=SFT Vision
2025.10
37.46
VisionThink-7B
2025.10
37.27
SFT Vision
2025.10
36.7
Qwen2.5-VL-Ins-7B
2025.10
36.42
Feedback
Search any
task
Search any
task