EASI Leaderboard

EASI: Holistic Evaluation of Multimodal LLMs on Spatial Intelligence

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-SI-1.1-Qwen2.5-VL-7B	58.816	52.51	42.6	70.87	50.36	62.25	60.07	76.01	84.29


SenseNova-SI-1.2-InternVL3-8B	64.5	69.6	42.6	89.0	58.8	49.0	60.1	69.4	77.7
Gemini 3 Pro	63.8	52.5	45.2	70.9	50.4	62.2	68.9	76.0	84.3
SenseNova-SI-1.1-InternVL3-8B	61.5	68.8	43.3	85.7	54.7	47.7	55.5	63.9	72.0
GPT-5	58.8	55.0	41.8	56.3	45.6	61.9	60.3	68.0	81.6
SenseNova-SI-1.1-Qwen3-VL-8B	58.1	64.8	38.1	73.8	51.2	49.6	53.2	61.9	72.5
Gemini 2.5 Pro	58.0	53.6	38.0	57.6	46.1	57.1	59.3	73.5	78.8
Seed 1.6	54.2	49.9	38.3	48.8	43.9	54.6	56.9	65.9	75.4
Grok 4	53.3	47.9	37.8	63.6	43.2	47.0	54.9	56.4	75.5
VST-7B-SFT	51.0	55.5	32.5	39.7	50.5	39.7	54.6	61.9	73.7
SenseNova-SI-1.1-Qwen2.5-VL-7B	51.0	58.1	32.8	54.7	45.5	43.9	46.3	55.3	71.4
Qwen3-VL-8B-Instruct	50.6	57.9	31.1	29.4	42.2	45.8	53.9	66.7	77.7
SenseNova-SI-1.1-InternVL3-2B	49.4	63.7	34.2	41.8	52.7	36.8	50.5	52.4	62.8
InternVL3_5-8B	49.0	56.1	29.0	40.2	40.0	43.8	49.2	58.2	75.7
VST-3B-SFT	48.4	51.4	28.8	36.0	52.9	35.9	54.1	58.8	69.0
vlm-3r-llava-qwen2-lora	46.6	60.7	27.9	40.0	40.5	31.3	51.5	52.3	68.2
Cambrian-S-7B	46.4	62.9	27.1	37.9	41.3	36.1	54.8	37.9	72.8
InternVL3-8B	45.7	42.1	28.0	41.5	38.7	41.1	44.2	53.5	76.3
SenseNova-SI-1.1-Qwen2.5-VL-3B	45.7	54.9	30.8	52.6	43.5	37.8	45.0	45.6	55.2
BAGEL-7B-MoT	45.3	31.4	31.0	34.7	41.3	37.0	50.2	63.6	73.1
Qwen3-VL-2B-Instruct	44.6	50.4	28.9	34.5	37.0	35.7	47.5	53.2	70.1
ViLaSR	43.7	44.6	30.2	35.1	35.7	38.7	46.6	51.4	67.3
Cambrian-S-3B	43.2	56.1	27.0	38.4	41.0	31.0	50.9	37.7	63.5
Qwen2.5-VL-7B-Instruct	42.6	32.3	26.8	36.0	36.9	37.6	43.5	55.9	71.8
SpaceR-SFT-7B	41.8	41.6	27.4	38.0	35.9	34.3	40.5	49.6	66.9
SpatialLadder-3B	40.9	44.9	27.4	43.5	39.9	28.0	42.8	43.0	58.2
Qwen2.5-VL-3B-Instruct	40.4	27.0	28.6	37.6	32.0	33.1	53.9	48.7	62.3
InternVL3-2B	39.8	33.0	26.5	37.5	32.6	30.0	47.7	50.8	60.1
Spatial-MLLM-subset-sft	35.6	46.3	26.1	33.5	34.7	18.0	36.2	40.5	50.0
MindCube-Qwen2.5VL-RawQA-SFT	22.0	17.2	1.7	51.7	24.1	6.3	2.8	35.1	37.0