LLM Persuasion Target Susceptibility
Safety · 2026-03-26
Measures how easy a model is to move as the target of persuasion dialogues. Lower scores indicate more resistance to opponents moving the model away from its prior position.
Top models (lower is better)
| Model | Score |
|---|---|
| Grok 4.20 | 0.0 |
| Kimi K2.5 | 0.4 |
| Opus 4.6 | 0.4 |
| Sonnet 4.6 | 0.6 |
| GPT-5.4 | 0.7 |
| Gemini 3.1 Flash-Lite Preview | 0.9 |
| ERNIE 5.0 | 1.0 |
| MiniMax M2.7 | 1.1 |
| Qwen3.5 397B A17B | 1.3 |
| Mistral Large 3 675B Instruct 2512 | 1.4 |
| GLM-5 | 1.6 |
| Doubao Seed 2.0 Pro | 1.7 |
| DeepSeek-V3.2 | 1.7 |
| Gemini 3.1 Pro Preview | 1.8 |
| MiMo-V2-Pro | 2.0 |