List your company

Safety Risk Assessment on EMBGUARD (test)

16.2Potential RiskInternVL3.5-1B
13.42432.16250.969.638May 29, 2026

Evaluation results

MethodLinks
InternVL3.5-1BCategory=Open-source2026.0516.214.69.8
Gemma-3-4BCategory=Open-source2026.0539.927.19.6
InternVL3.5-2BCategory=Open-source2026.0544.921.15.1
Gemma-3-27BCategory=Open-source2026.0545.363.716.4
Qwen-3-VL-30B-a3bCategory=Open-source2026.0546.16424.8
Qwen-3-VL-2BCategory=Open-source2026.0547.237.55.9
Qwen-3-VL-4BCategory=Open-source2026.0547.35110.5
Gemma-3-12BCategory=Open-source2026.0547.349.59.2
Qwen-3-VL-8BCategory=Open-source2026.0549.151.614.4
Qwen-3-VL-235B-a22bCategory=Open-source2026.0549.556.426.7
Qwen-3-VL-32BCategory=Open-source2026.0549.756.924.6
EMBGUARD-2BParameters=2B2026.0550.547.17.1
EMBGUARD-4BParameters=4B2026.0550.568.827.1
GPT-4o-miniCategory=Closed-source2026.0551.152.323.6
Qwen-3-VL-235B-a22bParameters=235B2026.0551.560.833.8
EMBGUARD-2BCategory=EMBGUARD2026.0551.644.67.4
GPT-4oCategory=Closed-source2026.0552.351.828.8
EMBGUARD-4BCategory=EMBGUARD2026.0554.350.314.6
GPT-5.12026.0555.54231.9
GPT-5.1Category=Closed-source2026.0555.858.133.4
Gemini-2.5-FlashCategory=Closed-source2026.0556.855.527
Gemini-2.5-ProCategory=Closed-source2026.0558.456.829.3
Human2026.0585.690.963.6