Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Spatial Reasoning on HR-Bench 4K
Loading...
77.5
HR-4K Average Score
Mini-o3
69.388
71.494
73.6
75.706
Apr 21, 2026
HR-4K Average Score
HR-4K Spatial Score (S)
HR-4K Context Score (C)
Updated 1mo ago
Evaluation Results
Method
Method
Links
HR-4K Average Score
HR-4K Spatial Score (S)
HR-4K Context Score (C)
Mini-o3
SFT=true, RL=true
2026.04
77.5
-
-
Simple o3
SFT=true, RL=false
2026.04
76.2
-
-
ToolsRL
SFT=false, RL=true
2026.04
75.9
91.2
60.6
DeepEyes
SFT=false, RL=true
2026.04
75.2
91.3
59
Pixel-Reasoner
SFT=true, RL=true
2026.04
74
-
-
Qwen2.5-VL
SFT=false, RL=false
2026.04
70.4
83.8
56.9
ZoomEye
SFT=false, RL=false
2026.04
69.7
84.3
55
Feedback
Search any
task
Search any
task