List your company

Safe Planning for Embodied Agents on EMBGUARD (test)

95.1Safe PrecisionGemini-2.5-Pro
90.00491.32792.6593.973May 29, 2026

Evaluation results

MethodLinks
Gemini-2.5-Pro2026.0595.142.833.454.9
Qwen-3-VL-32BModel Scale=32B2026.0592.46628.750
EMBGUARD-4BModel Scale=4B2026.0592.161.523.440.4
GPT-5.12026.0591.47135.634.5
EMBGUARD-2BModel Scale=2B2026.0590.2406.640.8