Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Perception on MMSTAR
Loading...
72.3
Accuracy
THINKLITE-VL
33.3
43.425
53.55
63.675
May 22, 2026
Accuracy
Updated 9d ago
Evaluation Results
Method
Method
Links
Accuracy
THINKLITE-VL
Method=THINKLITE-VL
2026.05
72.3
INTERNVL3-8B + PGT
Backbone=INTERNVL3-8B,...
2026.05
68.5
INTERNVL3-8B
Backbone=INTERNVL3-8B
2026.05
68.1
QWEN2.5-VL-7B + PGT
Backbone=QWEN2.5-VL-7B...
2026.05
63.5
IMAGE JIGSAW
Method=IMAGE JIGSAW
2026.05
62.9
QWEN2.5-VL-7B + SPECIALIZED MIX
Backbone=QWEN2.5-VL-7B...
2026.05
62.5
QWEN2.5-VL-7B
Backbone=QWEN2.5-VL-7B
2026.05
62.2
QWEN2.5-VL-3B + SPECIALIZED MIX
Backbone=QWEN2.5-VL-3B...
2026.05
56.9
QWEN2.5-VL-3B + PGT
Backbone=QWEN2.5-VL-3B...
2026.05
56.5
QWEN2.5-VL-3B
Backbone=QWEN2.5-VL-3B
2026.05
55.3
SPATIAL-LADDER-3B
Method=SPATIAL-LADDER-3B
2026.05
48.9
LLAVA-NEXT-LLAMA3-8B + PGT
Backbone=LLAVA-NEXT-LL...
2026.05
45.3
LLAVA-NEXT-LLAMA3-8B
Backbone=LLAVA-NEXT-LL...
2026.05
41.5
VIGORL-3B
Method=VIGORL-3B
2026.05
37.9
LLAVA-NEXT-7B + PGT
Backbone=LLAVA-NEXT-7B...
2026.05
37.1
LLAVA-NEXT-7B
Backbone=LLAVA-NEXT-7B
2026.05
34.8
Feedback
Search any
task
Search any
task