Artificial Analysis ships Capability Indices v1.1; Claude Fable 5.1 tops all six

ArtificialAnlys · x · 2026-09-15

Artificial Analysis releases Capability Indices v1.1, mapping ONET occupational tasks to weighted benchmarks across six indices: Finance & Accounting, Strategy & Ops, Legal, Healthcare, Engineering, and Economics. Key updates: Agentic Tool Use added from AutomationBench-AA, Agentic Customer Interaction removed, GDP.pdf added to Long-Context Reasoning, AA-Briefcase added to Agentic Knowledge Work, Terminal-Bench updated to v4.0 for Engineering. Results: Claude Fable 5.1 (max) leads all six indices; GPT-6 Astra (max) second in four. Open weights are competitive: Kimi K3 leads in Finance (#8), Legal (#9), Economics (#7); DeepSeek V4.1 Flash in Strategy & Ops (#7); GLM-5.3 in Healthcare (#6) and Engineering (#7).

Related event: Artificial Analysis Capability Indices v1.1: Claude Leads All Six Domains(3 posts)→

Original post →

More from Models

Models channel →