Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Factuality Detection on CN Text IF
Loading...
90.8
Score
Yuvion VL-32B
55.96
65.005
74.05
83.095
Jun 23, 2026
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
Yuvion VL-32B
Protocol=Reasoning
2026.06
90.8
Yuvion VL-8B
Protocol=Instruct
2026.06
89.6
Yuvion VL-8B
Protocol=Reasoning
2026.06
88.6
Yuvion VL-32B
Protocol=Instruct
2026.06
87.7
K2.5
Model Size=≈1T
2026.06
81.9
GLM-5
Model Size=≈700B
2026.06
79.7
Qwen3-VL-32B
Protocol=Instruct
2026.06
77.6
GPT-5.4
Model Size=≈1.5T
2026.06
77.3
Qwen3-VL-8B
Protocol=Instruct
2026.06
71.3
Qwen3-VL-32B
Protocol=Thinking
2026.06
69.4
Qwen3-VL-8B
Protocol=Thinking
2026.06
66.2
Qwen3.5-Plus
Model Size=397B
2026.06
57.3
Feedback
Search any
task
Search any
task