aiXamine - Fairness & Bias: leaderboard
Metric: Service score (0-100; normalized pass rate). Source: arxiv.org. 42 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B | 79 |
| 2 | Gemma 4 26B A4B | 77.5 |
| 3 | Gemini 3 Flash (Preview) | 76.7 |
| 4 | GPT-OSS-120B | 72.9 |
| 5 | Phi-4 | 72 |
| 6 | Qwen 3.5 27B | 71.2 |
| 7 | GPT-5 Mini | 68.8 |
| 8 | GPT-OSS-20B | 67.6 |
| 9 | GPT-4o | 67.5 |
| 10 | Gemini 2.0 Flash (001) | 66.9 |
| 11 | Gemma 4 E4B | 65.8 |
| 12 | Llama 3.1 8B | 65.7 |
| 13 | Gemma 2 27B | 64.7 |
| 14 | Qwen 3.5 122B A10B | 64.3 |
| 15 | DeepSeek R1 Distill Qwen 14B | 64 |
Interactive version: theaggregate.ai/benchmark?slug=aixamine-fairness-bias · How It Works · Data refreshed daily, snapshot 2026-09-19.