LLM Trustworthy - Out-of-Distribution — leaderboard
Metric: Trust Score (%). Source: huggingface.co. 26 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 (0314) | 87.55 |
| 2 | GPT-4o (2024-05-13) | 86.59 |
| 3 | Claude 2 | 85.77 |
| 4 | GPT-4o Mini (2024-07-18) | 79.07 |
| 5 | Llama 2 7B Chat | 75.65 |
| 6 | GPT-3.5 Turbo (0301) | 73.58 |
| 7 | Mistral-7B-OpenOrca | 73.4 |
| 8 | Llama 3 8B Instruct | 70.85 |
| 9 | Gemini 1.0 Pro | 70.85 |
| 10 | tulu-2-13B | 70.17 |
| 11 | tulu-2-7B | 69.3 |
| 12 | zephyr-7B-beta | 65.58 |
| 13 | mpt-7B-chat | 64.26 |
| 14 | gemma-7B (IT) | 61.78 |
| 15 | vicuna-7B-v1.3 | 59.1 |
Interactive version: theaggregate.ai/benchmark?slug=llm-trustworthy-out-of-distribution · How the rankings work · Data refreshed daily, snapshot 2026-07-22.