EASI Leaderboard

EASI: Holistic Evaluation and Analysis for Spatial Intelligence Made Easy

EASI conceptualizes a comprehensive taxonomy of spatial tasks that unifies existing benchmarks and a standardized protocol for the fair evaluation of state-of-the-art proprietary and open-source models.


sensenova/SenseNova-SI-1.1-InternVL3-8B	60.032	68.78	43.3	85.67	54.67	47.74


sensenova/SenseNova-SI-1.1-InternVL3-8B	60.0	68.8	43.3	85.7	54.7	47.7
Gemini 3 Pro	56.2	52.5	45.2	70.9	50.4	62.2
GPT-5	52.1	55.0	41.8	56.3	45.6	61.9
Gemini 2.5 Pro	50.5	53.6	38.0	57.6	46.1	57.1
Grok 4	47.9	47.9	37.8	63.6	43.2	47.0
Seed 1.6	47.1	49.9	38.3	48.8	43.9	54.6
sensenova/SenseNova-SI-1.1-InternVL3-2B	45.8	63.7	34.2	41.8	52.7	36.8
rayruiyang/VST-7B-SFT	43.6	55.5	32.5	39.7	50.5	39.7
Qwen/Qwen3-VL-8B-Instruct	41.3	57.9	31.1	29.4	42.2	45.8
nyu-visionx/Cambrian-S-7B	41.1	62.9	27.1	37.9	41.3	36.1
rayruiyang/VST-3B-SFT	41.0	51.4	28.8	36.0	52.9	35.9
nyu-visionx/Cambrian-S-3B	38.7	56.1	27.0	38.4	41.0	31.0
OpenGVLab/InternVL3-8B	38.3	42.1	28.0	41.5	38.7	41.1
Qwen/Qwen3-VL-2B-Instruct	37.3	50.4	28.9	34.5	37.0	35.7
inclusionAI/ViLaSR	36.9	44.6	30.2	35.1	35.7	38.7
hongxingli/SpatialLadder-3B	36.7	44.9	27.4	43.5	39.9	28.0
RUBBISHLIKE/SpaceR-SFT-7B	35.4	41.6	27.4	38.0	35.9	34.3
ByteDance-Seed/BAGEL-7B-MoT	35.1	31.4	31.0	34.7	41.3	37.0
Qwen/Qwen2.5-VL-7B-Instruct	33.9	32.3	26.8	36.0	36.9	37.6
OpenGVLab/InternVL3-2B	31.9	33.0	26.5	37.5	32.6	30.0
Diankun/Spatial-MLLM-subset-sft	31.7	46.3	26.1	33.5	34.7	18.0
Qwen/Qwen2.5-VL-3B-Instruct	31.7	27.0	28.6	37.6	32.0	33.1
MLL-Lab/MindCube-Qwen2.5VL-RawQA-SFT	20.2	17.2	1.7	51.7	24.1	6.3