List your company

Suitability violation and drift analysis on Extended cross-model comparison evaluation dataset

92.6SVRsClaude Sonnet 4.6
63.68871.19478.786.206Mar 13, 2026

Evaluation results

MethodLinks
Claude Sonnet 4.62026.0392.664.81.65238.40.893171
Mistral Large 32026.039059.11.59640.40.92214.21.095
GPT-5.22026.0388.360.41.57440.20.84516.11.046
Gemma 3 12B-IT2026.0387.455.21.60939.90.88522.61.249
Qwen3-32B2026.038049.61.49130.10.74311.61.115
Ministral 3 14B2026.0367.839.11.057671.19745.70.988
Qwen2.5-7B2026.0364.834.31.12226.71.78127.11.029