List your company

Harm Evaluation on PrincipalBench 36-item core

0Harm RateGemini-3p1-flash-lite
-3.1618.1739.560.83Jun 29, 2026

Evaluation results

MethodLinks
Gemini-3p1-flash-liteInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.060——
Mistral-LargeInteraction Arm=prompted, Evaluation Seeds=seed-12026.066——
Mistral-LargeInteraction Arm=plain, Evaluation Seeds=seed-12026.069——
DeepSeek-v3.1Interaction Arm=prompted, Evaluation Seeds=seed-12026.069——
DeepSeek-v3.1Interaction Arm=plain, Evaluation Seeds=seed-12026.0611——
DeepSeek-v3.1Interaction Arm=scaffolded, Evaluation Seeds=seed-12026.0611——
Claude-OpusInteraction Arm=plain, Evaluation Seeds=seed-12026.0611——
Llama-3.1-70BInteraction Arm=plain, Evaluation Seeds=seed-12026.0611——
Qwen3-32BInteraction Arm=prompted, Evaluation Seeds=seed-12026.0612——
Gemini-2.5-flashInteraction Arm=prompted, Evaluation Seeds=seed-12026.0614——
Qwen3-32BInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0614——
Gemini-3-flashInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0615——
Gemini-2.5-flashInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0617——
Gemini-3p1-flash-liteInteraction Arm=plain, Evaluation Seeds=seed-12026.0617——
Gemini-3p1-flash-liteInteraction Arm=prompted, Evaluation Seeds=seed-12026.0617——
Llama-3.1-70BInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0617——
Gemini-3-flashInteraction Arm=plain, Evaluation Seeds=seed-12026.0617——
Claude-SonnetInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0617——
Gemini-2.5-flashInteraction Arm=plain, Evaluation Seeds=seed-12026.0619——
Claude-OpusInteraction Arm=prompted, Evaluation Seeds=seed-12026.0619——
Claude-OpusInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0619——
Claude-SonnetInteraction Arm=plain, Evaluation Seeds=seed-12026.0619——
Gemini-3-flashInteraction Arm=prompted, Evaluation Seeds=seed-12026.0620——
Llama-3.1-70BInteraction Arm=prompted, Evaluation Seeds=seed-12026.0621——
Mistral-LargeInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0622——
Claude-SonnetInteraction Arm=prompted, Evaluation Seeds=seed-12026.0622——
Qwen3-32BInteraction Arm=plain, Evaluation Seeds=seed-12026.0625——
GLM-4.6Interaction Arm=plain, Evaluation Seeds=seed-12026.0643——
GLM-4.6Interaction Arm=scaffolded, Evaluation Seeds=seed-12026.0643——
GPT-5-miniInteraction Arm=plain, Evaluation Seeds=seed-12026.0644——
Qwen3.5-27BInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0659——
GLM-4.6Interaction Arm=prompted, Evaluation Seeds=seed-12026.0661——
GPT-5Interaction Arm=plain, Evaluation Seeds=seed-12026.0663——
GPT-5-miniInteraction Arm=scaffolded, Evaluation Seeds=seed-12026.0665——
GPT-5Interaction Arm=prompted, Evaluation Seeds=seed-12026.0671——
GPT-5Interaction Arm=scaffolded, Evaluation Seeds=seed-12026.0671——
Qwen3.5-27BInteraction Arm=plain, Evaluation Seeds=seed-12026.0672——
GPT-5-miniInteraction Arm=prompted, Evaluation Seeds=seed-12026.0676——
Qwen3.5-27BInteraction Arm=prompted, Evaluation Seeds=seed-12026.0679——
Claude-OpusEvaluation Seeds=n=52026.06—18.12.8
Claude-SonnetEvaluation Seeds=n=52026.06—19.51.5
DeepSeek-v3.1Evaluation Seeds=n=52026.06—12.32.5
Gemini-2.5-flashEvaluation Seeds=n=52026.06—5.56.3
Gemini-3-flashEvaluation Seeds=n=52026.06—19.42.3
Gemini-3p1-flash-liteEvaluation Seeds=n=52026.06—122.1
GLM-4.6Evaluation Seeds=n=52026.06—462.9
GPT-5Evaluation Seeds=n=52026.06—71.12.2
GPT-5-miniEvaluation Seeds=n=52026.06—53.65.1
Llama-3.1-70BEvaluation Seeds=n=52026.06—19.22.7
Mistral-LargeEvaluation Seeds=n=52026.06—112.8
Qwen3-32BEvaluation Seeds=n=52026.06—16.52.6
Qwen3.5-27BEvaluation Seeds=n=52026.06—75.32.9