Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Dense Multimodal Reasoning on Unifybench++
Loading...
85.4
GPT Score
GPT-4o+IP
29.136
43.743
58.35
72.957
May 11, 2026
GPT Score
BLEU
Updated 2mo ago
Evaluation Results
Method
Method
Links
GPT Score
BLEU
GPT-4o+IP
Model Size=200B, Metho...
2026.05
85.4
-
GPT-4o+TP
Model Size=200B, Metho...
2026.05
81.9
-
RAP-MLLM
Model Size=13B, Method...
2026.05
59.5
-
Yo’LLaVA
Model Size=13B, Method...
2026.05
52.9
-
MC-LLaVA
Model Size=13B, Method...
2026.05
51.1
-
Sync-R1
Model Size=1.3B, Metho...
2026.05
50.3
-
Yo’LLaVA(Phi-1.5)
Model Size=1.3B, Metho...
2026.05
48.4
-
Yo’Chameleon
Model Size=7B, Method...
2026.05
39.9
-
UniCTokens
Model Size=1.3B, Metho...
2026.05
38.5
-
Chameleon+IP
Model Size=7B, Method...
2026.05
37.4
-
Qwen2.5-VL + TP
Model Size=3B, Method...
2026.05
34.1
-
Show-o+TP
Model Size=1.3B, Metho...
2026.05
33.7
-
Chamaleon+TP
Model Size=7B, Method...
2026.05
31.3
-
Feedback
Search any
task
Search any
task