Psi-Bench - Viewpoint Debate Personalization: leaderboard
Metric: Personalized argumentation score (how well arguments are tailored to the hidden client profile), Change My View debates (1-9; DeepSeek-v3.2 judge, which also plays the persona-profiled client; three rounds after the client's opening message, client profile hidden from the model). Source: arxiv.org. Saturation forecast: Around January 2028. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 5.62 |
| 2 | GPT-5.1 | 5.57 |
| 3 | Gemini 3.1 Pro (Preview) | 5.54 |
| 4 | Grok 4 Fast | 5.2 |
| 5 | GPT-5 Mini | 4.91 |
| 6 | DeepSeek V3.2 | 4.8 |
| 7 | Gemini 3 Flash | 4.73 |
| 8 | Qwen 3 32B | 4.34 |
| 9 | Qwen 3 Next 80B A3B | 4.1 |
| 10 | Qwen 3 8B | 3.56 |
Interactive version: theaggregate.ai/benchmark?slug=psi-bench-viewpoint-debate-personalization · How It Works · Data refreshed daily, snapshot 2026-09-29.