Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Spatial Reasoning on MMSI
Loading...
32.3
Score
Qwen2.5-VL-7B + SAGE
23.876
26.063
28.25
30.437
Mar 11, 2026
Mar 22, 2026
Apr 2, 2026
Apr 14, 2026
Apr 25, 2026
May 6, 2026
May 18, 2026
Score
Updated 15d ago
Evaluation Results
Method
Method
Links
Score
Qwen2.5-VL-7B + SAGE
Model=Qwen2.5-VL-7B +...
2026.05
32.3
GPT-4o
Model=GPT-4o
2026.05
30.3
ViLASR-7B
FT-Data=73K, Fine-tune...
2026.03
30.2
SpatialLadder-3B
Model=SpatialLadder-3B
2026.05
29.2
Qwen2.5-VL-3B
FT-Data=-, Fine-tuned...
2026.03
28.6
VG-LLM
FT-Data=385k, Fine-tun...
2026.03
27.6
GeoSense
FT-Data=940K, Fine-tun...
2026.03
27.5
SpaceR-sft-7B
FT-Data=151K, Fine-tun...
2026.03
27.4
SpatialLadder-3B
FT-Data=26K, Fine-tune...
2026.03
27.4
Cambrain-S-3B
FT-Data=10M, Fine-tune...
2026.03
27
Qwen2.5-VL-7B
FT-Data=-, Fine-tuned...
2026.03
26.8
InternVL3-2B
FT-Data=-, Fine-tuned...
2026.03
26.5
Qwen2.5-VL-7B
Model=Qwen2.5-VL-7B
2026.05
26.4
InternVL-2.5-8B
Model=InternVL-2.5-8B
2026.05
25.7
VG-LLM*
FT-Data=940k, Fine-tun...
2026.03
25.2
LLaVA-OV-7B
Model=LLaVA-OV-7B
2026.05
24.5
Qwen2.5-VL-3B*
FT-Data=940k, Fine-tun...
2026.03
24.2
Feedback
Search any
task
Search any
task