Benchmark
Persuasion – Target Susceptibility (Lech Mazur)
Scores are percentages.
Benchmark sourceScores from Lech Mazur suite data.
- Models measured
- 14
- Scores recorded
- 14
- Sources
- 1
- IQ weight
- 0.00
Score distribution
- 1.1%
- Mean
- 1.1%
- Median
- 0.578%
- Std dev
- 0.367%
- 10th percentile
- 1.8%
- 90th percentile
- 0.015%
- Min
This benchmark has a weight of 0.00 for chat models, measured over 14 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | MiMo V2 Pro· Lech Mazur suite data | 2% |
| 2 | Gemini 3.1 Pro Preview· Lech Mazur suite data | 1.8% |
| 3 | DeepSeek V3.2· Lech Mazur suite data | 1.7% |
| 4 | GLM 5· Lech Mazur suite data | 1.6% |
| 5 | Mistral Large 3· Lech Mazur suite data | 1.4% |
| 6 | Qwen3.5 397B A17B· Lech Mazur suite data | 1.3% |
| 7 | MiniMax M2.7· Lech Mazur suite data | 1.1% |
| 8 | ERNIE 5.0· Lech Mazur suite data | 1% |
| 9 | Gemini 3.1 Flash Lite· Lech Mazur suite data | 0.898% |
| 10 | GPT 5.4high reasoning· Lech Mazur suite data | 0.724% |
| 11 | Claude Sonnet 4.6high reasoning· Lech Mazur suite data | 0.613% |
| 12 | Claude Opus 4.6high reasoning· Lech Mazur suite data | 0.407% |
| 13 | Kimi K2.5thinking· Lech Mazur suite data | 0.367% |
| 14 | Grok 4.20Reasoning· Lech Mazur suite data | 0.015% |