Polar (US) - Sociocultural Axis: leaderboard
The sociocultural axis of Polar (2026) on the U.S. set in English: an ICAT score from 0 to 100 combining language-modelling competence with balance between progressive and conservative positions, from option likelihoods. Higher is better.
Metric: Sociocultural-axis ICAT (0-100), the StereoSet idealized context association score adapted to political bias: the language-model score LMS (share of items where the unrelated continuation is not chosen) times the neutrality score NS (how close the preference between the left or progressive and the right or conservative continuation is to balance, averaged over issue categories); options chosen by length-normalized log-likelihood in LM Evaluation Harness; higher means more balanced and competent; U.S. political statements in English. Source: arxiv.org. 38 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.2 1B | 87.31 | #9258 |
| 2 | Qwen3-8B-Base | 86.67 | #1721 |
| 3 | Llama 3.1 70B | 85.35 | #1490 |
| 4 | Llama 3.2 3B | 84.66 | #7830 |
| 5 | Qwen3-1.7B-Base | 84.29 | #4890 |
| 6 | Midm-2.0-Base-Instruct | 84.25 | #1921 |
| 7 | Qwen 3 32B | 83.84 | #960 |
| 8 | Mistral-Small-24B-Base-2501 | 83.64 | #1450 |
| 9 | Llama 3.1 8B | 82.68 | #4861 |
| 10 | Qwen3-0.6B-Base | 82.47 | #8156 |
| 11 | Llama 3.2 1B Instruct | 82.36 | #9047 |
| 12 | Qwen 3 8B | 82.33 | #1830 |
| 13 | Qwen3-14B-Base | 82.12 | #1504 |
| 14 | Qwen3-4B-Base | 81.57 | #2515 |
| 15 | Qwen 3 4B | 81.46 | #2281 |
Interactive version: theaggregate.ai/benchmark?slug=polar-us-sociocultural-axis · How It Works · Data refreshed daily, snapshot 2026-11-13.