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

Spatial Reasoning on ViewSpatial-Bench

51.8Overall Score

MSSR

Updated 1mo ago

Evaluation Results

Method	Links
MSSR 2025.10		51.8	51	54.4
o3 2025.10		51.1	51.3	51
LEO 2025.10		43.7	41.5	45.8
InternVL2.5-8B 2025.10		43.2	46.5	40.2
Qwen2.5-VL-72B 2025.10		43.1	47.6	38.9
Gemini 2.5 Pro 2025.10		43	44.3	41.6
Qwen3-VL-8B 2025.10		41.2	45.1	37.5
InternVL3-14B 2025.10		40.3	47.1	33.9
V2LO-7B 2025.11		40.18	41.66	38.78
VLM-3R 2025.10		39.2	43.8	34.9
Qwen3-VL-4B 2025.10		39	42.3	35.9
Gemini 2.5 Flash 2025.10		38.4	41.6	36.9
SpaceR-7B 2025.11		38.38	41.1	35.81
ViLaSR 2025.10		38.2	42.4	34.4
InternVL2.5-8B 2025.11		38.13	36.8	39.38
V2LO-7B 2025.11		38.13	39.13	36.83
Qwen2.5VL-7B 2025.11		37.11	39.54	34.83
Qwen2.5-VL-7B 2025.10		36.9	40.6	33.4
Qwen2.5-VL-3B 2025.10		35.8	39.8	32.1
GPT-4o 2025.10		35	33.6	36.3
GPT-4o 2025.11		34.95	-	-
LLaVA-3D 2025.10		33.8	39.6	28.3
VADAR 2025.10		33.7	34.2	33.2
Gemini 2.0 flash 2025.11		32.56	-	-
Video 3D LLM 2025.10		30.3	24.4	35.8
Llama-3.2-11B-Vision 2025.10		28.8	23.7	33.6
LLaVA-OneVision-7B 2025.10		27.5	28.5	26.6
LLaVA-OneVision-7B 2025.11		27.49	28.49	26.54