List your company

Causal Spatial Reasoning on CausalSpatial

84.49Overall Average ScoreHuman
26.364441.454756.54571.6353Jan 19, 2026

Evaluation results

MethodLinks
Human2026.0184.49—
GPT52026.0154.171
Gemini2.5 Pro2026.0150.552
Qwen3-VL-30B-A3B-T.2026.0150.53
Gemini2.5 Flash2026.0148.824
Sonnet 3.72026.0148.075
Qwen3-VL-30B-A3B-I.2026.0145.86
GPT5-mini2026.0145.447
Qwen3-VL-4B-I.2026.0144.768
Qwen3-VL-8B-I.2026.0143.539
Qwen3-VL-8B-T.2026.0142.6510
Qwen3-VL-2B-I.2026.0140.611
Spatial-MLLM2026.014012
Qwen2.5-VL-7B-I.2026.0139.6113
LLaVAOneVision1.5-8B-I.2026.0134.6214
Spatial-VLM2026.0133.515
Random2026.0128.6—