List your company

Safety Evaluation on Harmful Prompts Text-only baseline

0Text ASRclaude-sonnet-4.5-20250929
-3.63220.88445.469.916May 8, 2026

Evaluation results

MethodLinks
claude-sonnet-4.5-20250929Model Type=Proprietary2026.050
claude-haiku-4.5-20251001Model Type=Proprietary2026.058
gpt-4.1-2025-04-14Model Type=Proprietary2026.0512.7
doubao-seed-1.6-251015Model Type=Proprietary2026.0517.3
glm-4.5vModel Type=Open Source2026.0529.2
qwen3-vl-235b-a22b-thinkingModel Type=Open Source2026.0531.4
qwen3-vl-plusModel Type=Proprietary2026.0535.3
qwen3-vl-32b-thinkingModel Type=Open Source2026.0536.7
gemini-2.5-flashModel Type=Proprietary2026.0547.5
intern-s1Model Type=Open Source2026.0590.8