← October 28, 2026 briefing · added in the 16:00 KST update
Artificial Analysis overhauls Intelligence Index to v5: open-closed gap widens from 3 to 12 points, 60% from private test sets (community report)
Announced October 28, 2026
Verified (fact)
According to a Reddit post, Artificial Analysis has overhauled its Intelligence Index to v5. According to the announcement cited in the post, the overhaul widened the score gap between open-weight and closed models from 3 points to 12 points. Sixty percent of the index comes from private test sets or evaluations with unpublished answers, so results cannot be reproduced externally. Some also pointed out that rankings such as 'GPT-6.1 Terra ranks above Fable 5.1' do not match real-world experience.
Needs fact-check (claim)
Artificial Analysis says v5 ranks models more accurately.
Why it matters (interpretation)
Because the index is often cited in model selection and media coverage, the overhaul alone could significantly change the view that 'open models have caught up.' At the same time, the larger share of private evaluations makes it harder to verify.
Confidence medium · official source pending
Sources
- Community Artificial Analysis Intelligence Index v5 (major update) Reddit
- Community Artificial Analysis Intelligence Index v5 released Reddit
More about Artificial Analysis
-
Chinese models lead Artificial Analysis video-with-audio ranking: Alibaba's Wan 3.0 first, MiniMax H3 third, Seedance 2.5 fourth (report)
According to AI Times, Alibaba's 'Wan 3.0' ranked first with 1,175 points in Artificial Analysis's text-to-video-with-audio evaluation...
-
Mistral Large 4 follow-up: an improved release with more RL after preview, pitched for cyber defense, scores 48% on Coding Agent Index (reported)
According to TechCrunch, Mistral AI released the weights of Mistral Large 4, a 1T-parameter model, and pitched cybersecurity as a main...
-
DeepSeek V4.1-Flash-1017 follow-up: independent Terminal-Bench 4.0 scores of 34% and 31.8% trail the model card's 42.3% (community reports)
According to a Reddit post, the model card lists V4.1-Flash-1017's Terminal-Bench 4.0 score as 42.3%. On the same benchmark, Artificial...
-
Artificial Analysis releases 'AA-Robotics v0.1', a benchmark of how well LLMs perform robotics tasks (community report)
According to a Reddit post, Artificial Analysis released AA-Robotics v0.1, a benchmark that ranks LLMs on robotics task performance....