Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
MM STEM & Puzzle on MathVista mini
Loading...
89.1
Score
K2.5
76.204
79.552
82.9
86.248
Jun 23, 2026
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
K2.5
Parameter Count=≈1T
2026.06
89.1
Qwen3.5-Plus
Parameter Count=397B
2026.06
86.7
Qwen3-VL-32B
Decoding Mode=Thinking...
2026.06
83.9
Qwen3-VL-32B
Decoding Mode=Instruct...
2026.06
82.4
Yuvion VL-32B
Decoding Mode=Reasonin...
2026.06
81.4
Yuvion VL-32B
Decoding Mode=Instruct...
2026.06
80.5
Qwen3-VL-8B
Decoding Mode=Thinking...
2026.06
79.6
Yuvion VL-8B
Decoding Mode=Reasonin...
2026.06
79.2
Opus-4.6
Parameter Count=≈800B
2026.06
77.6
GPT-5.4
Parameter Count=≈1.5T
2026.06
77.2
Yuvion VL-8B
Decoding Mode=Instruct...
2026.06
77
Qwen3-VL-8B
Decoding Mode=Instruct...
2026.06
76.7
Feedback
Search any
task
Search any
task