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

Spatial Reasoning on ReVSI

60.9Average Score

Gemini 3 Pro

Updated 1mo ago

Evaluation Results

Method
Gemini 3 Pro 2026.05	60.9	60.1	54.7	79.3	51.9	68.1	56	56.4
Gemini 3 Pro 2026.06	60.9	60.1	54.7	79.3	51.9	68.1	56	56.4
S-Agent 2026.06	58.8	54	45.6	62.6	53.4	63.6	66.4	66.1
Gemini 3 Flash 2026.05	57.6	65.7	53.1	77.6	52.8	64.6	47.9	41.8
LLaVA-OV-2 2026.05	57.6	-	-	-	-	-	-	-
Gemini 3 Flash 2026.06	57.6	65.7	53.1	77.6	52.8	64.6	47.9	41.8
GeoWeaver-10B 2026.05	57.5	42.6	66.8	69.8	50.9	64.4	52.4	55.4
GeoWeaver-10B 2026.05	57.4	40.5	64.5	69.3	-	62.1	50.5	-
GeoWeaver-10B 2026.05	57.3	43.2	66.5	69.7	50.7	65	51.7	54.7
GeoWeaver-5B 2026.05	54.9	38	63.4	63.6	54.2	65.3	46.9	53.1
GeoWeaver-5B 2026.05	54.2	38.6	61.1	62.8	-	61.4	47	-
GeoWeaver-5B 2026.05	54.1	38.6	61.3	62.2	53.7	63.8	46.1	53.1
InternVL3.5-38B 2026.06	54.1	43.8	60.6	70.2	58.4	57.4	45.9	42.7
Qwen3-VL-32B-Instruct 2026.06	53.3	46.9	65	70.4	55.8	53.8	34	47.3
Cambrian-P 2026.05	52	41.4	68.3	66.7	45.9	40.2	48.7	53.1
GPT-5.2 2026.05	50.9	56.2	41.5	73.9	63	48.4	34.9	38.2
GPT-5.2 2026.06	50.9	56.2	41.5	73.9	63	48.4	34.9	38.2
VLM-3R-7B 2026.05	50.1	41.6	61.6	64.8	52.5	46.5	49.5	34.1
Cambrian-P (w/o Pose) 2026.05	50.1	42.3	64.6	64.5	47.4	38.1	48.3	45.3
VLM3R-7B 2026.06	50.1	41.6	61.6	64.8	52.5	46.5	49.5	34.1
Cambrian-S-7B 2026.05	49.1	48.4	60.5	65.5	46.7	37.1	48.5	37
Qwen3-VL-8B-Instruct 2026.05	49.1	40.4	52.3	69	45.1	57.1	39.5	40.5
Cambrian-S-7B 2026.05	49.1	48.4	60.5	65.5	46.7	37.1	48.5	37
Qwen3-VL-8B-Instruct 2026.06	49.1	40.4	52.3	69	45.1	57.1	39.5	40.5
Cambrian-S-7B 2026.06	49.1	48.4	60.5	65.5	46.7	37.1	48.5	37
Qwen3-VL 2026.05	48.9	-	-	-	-	-	-	-
InternVL3.5-8B 2026.05	47.9	43.3	54.6	64.2	47.6	45	36.3	44.4
InternVL-3.5 2026.05	47.9	-	-	-	-	-	-	-
InternVL3.5-8B 2026.06	47.9	43.3	54.6	64.2	47.6	45	36.3	44.4
VST-7B-SFT 2026.05	46.4	35.4	52.6	67.9	47.2	49.2	36.9	35.4
VST-7B-SFT 2026.05	46.4	35.4	52.6	67.9	47.2	49.2	36.9	35.4
VST-7B-SFT 2026.06	46.4	35.4	52.6	67.9	47.2	49.2	36.9	35.4
Spatial-MLLM-4B-820k 2026.05	40.9	41.5	40	53.1	-	30.7	39.2	-
Spatial-MLLM-4B-820k 2026.06	40.9	41.5	40	53.1	-	30.7	39.2	-
Spatial-MLLM-4B-135k 2026.05	40.5	40.7	45.3	46.8	-	32.3	37.4	-
Spatial-MLLM-4B-135k 2026.06	40.5	40.7	45.3	46.8	-	32.3	37.4	-
LLaVA-Video-72B-Qwen2 2026.06	37.8	40.1	29.6	59.3	27.9	39.6	24.8	43
Qwen2.5-VL-7B-Instruct 2026.06	33.9	34.3	21.7	45.5	35.1	32.6	33.7	34.1
LLaVA-OV-1.5 2026.05	33.5	-	-	-	-	-	-	-
Qwen2.5-VL-7B-Instruct 2026.05	32.6	36.9	15	49.7	29	31.5	29.5	36.7
Qwen2.5-VL-7B-Instruct 2026.06	32.6	36.9	15	49.7	29	31.5	29.5	36.7
Keye-VL-1.5 2026.05	32.4	-	-	-	-	-	-	-
Chance (Frequency) 2026.06	31.4	52.2	40.1	17.4	20.9	25.8	31.9	30.2
LLaVA-Video-7B-Qwen2 2026.06	30.8	29.9	1.5	53	19.3	39.1	33.8	38.8
PLM 2026.05	30.7	-	-	-	-	-	-	-
SpaceR-7B 2026.05	30.5	30.7	34.5	52	18.6	22.8	34.5	20.2
SpaceR-7B (SG-RLVR) 2026.06	30.5	30.7	34.5	52	18.6	22.8	34.5	20.2
LLaVA-Video-7B-Qwen2 2026.05	30.3	31.3	1.4	52.5	16.7	38.3	33.3	38.4
LLaVA-Video-7B-Qwen2 2026.06	30.3	31.3	1.4	52.5	16.7	38.3	33.3	38.4
Qwen2.5-VL-3B-Instruct 2026.06	23.7	18.7	15.6	16.8	-	33.2	34.3	-
Chance (Random) 2026.06	-	-	-	-	-	23.7	26.8	26