LLM Trustworthy - Fairness: leaderboard
The "Fairness" subtask of LLM Trustworthy. AI-Secure's LLM trustworthiness leaderboard (DecodingTrust) — safety and reliability measured across adversarial robustness, ethics, fairness, out-of-distribution robustness, privacy, stereotype and toxicity dimensions, each reported as...
Metric: Trust Score (%). Source: huggingface.co. 26 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 2 7B Chat | 100 | #1499 |
| 2 | falcon-7B Instruct | 100 | #1622 |
| 3 | RedPajama-INCITE-7B-Instruct | 100 | #1566 |
| 4 | mpt-7B-chat | 100 | #1556 |
| 5 | tulu-2-13B | 97.9 | #1404 |
| 6 | Claude 2 | 96.81 | #761 |
| 7 | zephyr-7B-beta | 95.07 | #1330 |
| 8 | gemma-7B (IT) | 93.88 | #1496 |
| 9 | gemma-2B (IT) | 93.02 | #1559 |
| 10 | vicuna-7B-v1.3 | 85.53 | #1507 |
| 11 | tulu-2-7B | 83.21 | #1429 |
| 12 | Llama 3 8B Instruct | 80.05 | #1129 |
| 13 | Gemini 1.0 Pro | 80.05 | #933 |
| 14 | GPT-4o Mini (2024-07-18) | 69.74 | #558 |
| 15 | Mistral-7B-OpenOrca | 66.76 | #1389 |
Interactive version: theaggregate.ai/benchmark?slug=llm-trustworthy-fairness · How It Works · Data refreshed daily, snapshot 2026-10-20.