Artificial Analysis has released version 4.2 of its Intelligence Index, introducing new evaluations like AA-Briefcase for agentic knowledge work and GDP.pdf for long-context document reasoning. This update increases the weighting of private, held-out test sets to 40% to prevent gaming and includes infrastructure upgrades for more robust grading. The latest rankings show Anthropic's Claude Fable 5.1 leading, followed by OpenAI's GPT-6 Astra and Meta. AI
IMPACT This update to a prominent AI benchmark index reflects the rapid pace of model development and aims to provide more realistic evaluations for agentic capabilities and long-context reasoning.
RANK_REASON Update to an existing AI benchmark index with new evaluations and updated rankings.
Read on Hacker News — AI stories ≥50 points →
- AA-Briefcase
- AA-LCR v1.1
- AA-Omniscience
- Anthropic
- Artificial Analysis Intelligence Index v4.2
- Claude Fable 5.1
- CritPt
- GDP.pdf
- GDPval-AA v2
- GPQA Diamond
- GPT-5.6 Sol
- GPT-6 Astra
- Grok 4.5
- Meta
- Moonshot/Kimi
- OpenAI
- SpaceXAI
- Z.AI
- SciCode
- Surge AI
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →