The Intelligence Index v5 from Artificial Analysis was announced on October 27, 2026, the biggest update since v4 in January. For the first time, 60% of the index is based on evaluations with private test sets or held-out answers, to reduce contamination from training data.
What changed
New benchmarks include FrontierCode 2.0 (real coding tasks in open-source repositories, graded on whether a maintainer would merge the change) and Terminal-Bench-Science (70 agentic scientific research tasks in a terminal). Upgrades to HLE-Diamond (replacing Humanity's Last Exam), AA-Omniscience v1.1 and CritPt v2. Staying: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0. Removed: GDPval-AA v2.1, SciCode and AA-LCR v1.1.
Weights are 25% each for Knowledge Work, Coding, Science and General.
Key results
Claude Opus 5.5 (max) leads with 58 points, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55). Anthropic and OpenAI hold the top five. Gemini 4 Argon (high) is next at 50.
The open-weights gap widened from 3 points previously to 12. Kimi K3.1 Pro (weights not yet released) is the top open model at 46. Among released weights, DeepSeek V4.1 Flash 1017 and GLM-5.3 lead at 32 — 26 points behind the leader and below Claude Haiku 5.5 (36).
Cost per task varies by more than 13x among leaders. GPT-6.1 Terra (max) is cheapest at $0.96 per index task; Claude Fable 5.1 (max) reaches $13.04.
Why it matters
The index reflects current real-world usage (knowledge work, coding, science and general reasoning). The larger private share helps measure genuine capability rather than memorization. No single model leads all nine evaluations: Opus 5.5 leads three (FrontierCode 2.0, HLE-Diamond and AA-Omniscience), GPT-6 Astra two, and others the rest.
Artificial Analysis also launched Progress Index v1, which places models since November 2022 on one continuous capability scale.
Sources
Image credit: Artificial Analysis
Transparency: This content was created, edited or reviewed with the aid of artificial intelligence. Information was cross-checked with public posts on X and sources available on the internet. Consult the original sources to verify the full context.
By GeekikiBot
.avif)