Syco-bench - Mirroring: leaderboard
The model gives its view after the user takes one stance and then the opposite one, with no system prompt; the number is the mean absolute shift between the two judged positions (0-10). Measures stance mirroring, not factual capitulation.
Metric: Mean absolute shift in the model's stance between the two opposite user stances (-5..5 scale, so 0-10). Source: www.syco-bench.com. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.7 Sonnet (Thinking) | 1.18 | #458 (Claude 3.7 Sonnet) |
| 2 | Claude 3.5 Haiku | 1.35 | #1553 |
| 3 | Grok 3 Mini Beta | 1.58 | #616 |
| 4 | Gemini 2.5 Pro (Preview) | 1.6 | #253 |
| 5 | Claude 3.7 Sonnet | 1.63 | #458 |
| 6 | Grok 3 | 1.8 | #602 |
| 7 | O3 | 1.88 | #228 |
| 8 | Claude 3 Opus | 1.93 | #1455 |
| 9 | O4 Mini | 2.08 | #353 |
| 10 | Gemini 2.5 Flash (Preview) | 2.3 | #785 |
| 11 | GPT-4o Mini | 2.33 | #1623 |
| 12 | GPT-4o ChatGPT | 2.78 | #466 |
Interactive version: theaggregate.ai/benchmark?slug=syco-bench-mirroring · How It Works · Data refreshed daily, snapshot 2026-11-13.