Understand this story
Artificial Analysis released observations on October 23, 2026, regarding the relationship between Slop and intelligence level in AA-PersonalityBench. It stated that models with lower intelligence levels generally produce more Slop, but the differences in Slop among frontier models remain large. The observation gave as an example that GPT-6 Sol and Muse Spark 1.3 both have an Intelligence Index of 48, yet their Slop rankings are 1st and 80th, respectively; Muse Spark 1.3 (Slop 80), Qwen3.8 Max (75), and Gemini 4 Argon (70) belong to the category of high intelligence but also high Slop, while GPT-6 Luna (4) and Claude Haiku 5.5 (9) have relatively low Slop for their intelligence levels. There have been no subsequent developments so far.
AI-generated from public reports
Latest development · Report
AA-PersonalityBench: Significant Slop Differences Among Frontier ModelsDevelopments
1 confirmed developments
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 23
Artificial Analysis
AA-PersonalityBench: Significant Slop Differences Among Frontier Models
Artificial Analysis's AA-PersonalityBench shows that models with lower intelligence levels generally produce more Slop, but the Slop differences among frontier models remain large. GPT-6 Sol and Muse Spark 1.3 share the same Intelligence Index of 48, yet their Slop rankings are 1st and 80th respectively; Muse Spark 1.3 (Slop 80), Qwen3.8 Max (75) and Gemini 4 Argon (70) belong to the category of high intelligence but also high Slop, while GPT-6 Luna (4) and Claude Haiku 5.5 (9) have relatively low Slop for their intelligence levels.