List your company

Multimodal Physical Reasoning on PACC (test)

2.286Average ScoreGemini 2.5 Flash
1.026561.353531.68052.00747May 6, 2026

Evaluation results

MethodLinks
Gemini 2.5 FlashModel Access Type=Proprietary2026.052.2862.4692.4412.8332.32.5711.8751.8182.2
GPT-4oModel Access Type=Proprietary2026.052.122.1562.3242.6672.52.0481.6792.252.1
PhyARModel Access Type=Open-Source2026.052.0362.3752.2941.6112.051.6191.7322.5232.95
Qwen3.5-35BModel Access Type=Open-Source, Model Parameters=35B2026.051.4771.1251.5291.51.951.5241.4291.4091.55
VideoLLaMA3-7BModel Access Type=Open-Source, Model Parameters=7B2026.051.23711.1181.2221.21.2621.2681.3861.35
MiniCPM-V 2.6Model Access Type=Open-Source2026.051.1821.1251.1181.3331.451.1671.1611.1591.15
VideoLLaMA2-7BModel Access Type=Open-Source, Model Parameters=7B2026.051.15911.1181.1111.251.2381.0711.2041.25
InternVL2.5-8BModel Access Type=Open-Source, Model Parameters=8B2026.051.11711.14711.31.1311.1431.0911.1
Flash-VStream-7BModel Access Type=Open-Source, Model Parameters=7B2026.051.1011.1251.1761.22211.0241.1071.1821.05
Video-LLaVA-7BModel Access Type=Open-Source, Model Parameters=7B2026.051.08111111.0951.1431.1361.15
Video-ChatGPT-7BModel Access Type=Open-Source, Model Parameters=7B2026.051.07511.2061.0561.11.0481.0891.0451.1