LLM Trustworthy - Toxicity — leaderboard

Metric: Trust Score (%). Source: huggingface.co. 26 models tracked.

Top models

#ModelScore
1Claude 292.11
2GPT-4o (2024-05-13)86.46
3Llama 2 7B Chat80
4Llama 3 8B Instruct77.53
5Gemini 1.0 Pro77.53
6gemma-2B (IT)77.07
7gemma-7B (IT)75.52
8GPT-4o Mini (2024-07-18)59.02
9GPT-3.5 Turbo (0301)47
10tulu-2-13B44.8
11GPT-4 (0314)41
12mpt-7B-chat40
13falcon-7B Instruct39
14zephyr-7B-beta31.97
15Mistral-7B-OpenOrca30.12

Interactive version: theaggregate.ai/benchmark?slug=llm-trustworthy-toxicity · How the rankings work · Data refreshed daily, snapshot 2026-07-22.