27 AI models were ranked by the public and ChatGPT came 8th — these are the models that beat it

Abstraction: Prolific's Humaine leaderboard ranks Gemini first in public user experience study

Key points:

• Prolific's Humaine leaderboard evaluated 27 AI models via 21,352 pairwise user experience comparisons (not task-completion benchmarks); methodology based on users choosing which model performed better in side-by-side interactions
• Gemini 2.5 Pro ranked first overall across nearly all demographic filters: age groups in UK and US, political affiliation, ethnicity
• Rankings 2-4: Deepseek, Mistral Le Chat, Grok; Le Chat is noted for loyal fanbase despite lower benchmark scores
• ChatGPT (GPT-4.1) placed 8th; Claude's two v4 models placed 11th and 12th
• Grok-3 ranked highest for trust, ethics, and safety — described as ironic given Grok's controversies
• User experience rankings diverge significantly from standard benchmark results; highlights the gap between capability benchmarks and real-world perceived quality

Connections: Gemini · Chatgpt · Grok · Deepseek · Anthropic · Large Language Models · AI Benchmarking

Source: https://www.tomsguide.com/ai/27-ai-models-were-ranked-by-the-public-and-chatgpt-came-8th-these-are-the-models-that-beat-it