List your company

Tool-use security and utility evaluation on Security

98.4Utility Rategpt-5.4-mini
2.92827.71452.577.286May 19, 2026

Evaluation results

MethodLinks
gpt-5.4-miniTier=frontier, Mode=aware2026.0598.40.50.4089.9
Gemma-4-26B-A4B-itTier=open, Mode=aware2026.0595.33.60.466.773.6
Claude Sonnet 4.6Tier=frontier, Mode=aware2026.0578.55.21.666.777.8
Gemma-4-26B-A4B-itTier=open, Mode=blind2026.0572.96.20057
gpt-oss-20BTier=open, Mode=aware2026.0571.426.6010054.5
Gemini 3.1 Flash LiteTier=frontier, Mode=aware2026.0569.814.8010059.6
gpt-oss-20BTier=open, Mode=blind2026.0560.32.50046.1
Ministral-3-14BTier=open, Mode=blind2026.0557.84.20050.4
Gemini 3.1 Flash LiteTier=frontier, Mode=blind2026.0553.232.1010046.5
Qwen3.6-35B-A3BTier=open, Mode=blind2026.0548.440.11.566.742.2
Qwen3.6-35B-A3BTier=open, Mode=aware2026.0540.158.31.266.737.6
gpt-5.4-miniTier=frontier, Mode=blind2026.0528.154.20.810032.2
Claude Haiku 4.5Tier=frontier, Mode=aware2026.0527.967.9010039.5
Claude Sonnet 4.6Tier=frontier, Mode=blind2026.0511.277.111.910018.6
Ministral-3-14BTier=open, Mode=aware2026.057.888.5010014.3
Claude Haiku 4.5Tier=frontier, Mode=blind2026.056.684.12.810014.9