Share your thoughts, 1 month free Claude Pro on usSee more

Spatial Reasoning on CV-Bench (Component Metrics)

86.5Overall Score

SpatialStack

Updated 5d ago

Evaluation Results

Method	Links
SpatialStack 2026.03		86.5	69	92.5	93.7	90.7	-	-	-	-
GASP - Full Model (Lcorr + Ldepth) 2026.05		79.8	-	-	-	-	68.2	88.3	87.3	79.2
Qwen2.5-VL-32B 2026.05		79.7	-	-	-	-	68.9	80.8	86.5	85.8
LLaVA-OneVision-72B 2026.05		79.7	-	-	-	-	70.2	89.2	82.5	79
GASP - Correspondence (Lcorr) 2026.05		79.4	-	-	-	-	68	88.1	86.6	78.6
Qwen2.5-VL-7B-Instruct 2026.05		76.6	-	-	-	-	63.7	87.7	85.5	72.7
InternVL2.5-4B 2026.05		74.1	-	-	-	-	68	79.9	80.7	69.2
LLaVA-OneVision-7B 2026.05		73.2	-	-	-	-	69.2	77.9	81.7	65
LLaMA-3.2V-11B-CoT 2026.05		72.8	-	-	-	-	59.1	78.9	78.8	78
SpatialRGPT 2026.03		72.7	60.4	78.9	80	71.3	-	-	-	-
Qwen2.5-VL-3B-Instruct 2026.05		68.5	-	-	-	-	62.6	70.3	78	64.7
Spatialbot 2026.03		68	61.4	73.1	76.5	61	-	-	-	-
Cambrain-8B 2026.05		62.2	-	-	-	-	60.7	81.7	55	50.5
LLaMA-3.2V-11B 2026.05		58.2	-	-	-	-	59	55.9	67.3	50.5
LLaVA-1.5-13B 2026.05		58.2	-	-	-	-	56.1	57.2	66.8	53.3
SpaceLLaVA-13B 2026.05		58.2	-	-	-	-	56.1	57.2	66.8	53.3
Kimi-VL-3B-Thinking 2026.05		57.5	-	-	-	-	60.5	79.1	43.5	44
SpaceQwen2.5VL-3B 2026.05		51.4	-	-	-	-	62.2	45.4	45.4	50