New Agentic Benchmark: Claude Fable and GLM 5.2 Lead in Task Execution

Core summary

One sentence to understand this update

A new agentic benchmark from Artificial Analysis shows Claude Fable and GLM 5.2 leading their cohorts in LLM task planning and execution capabilities.

Impact & opportunity

What this could mean

This benchmark offers builders a new reference for evaluating and selecting agent models, helping assess their performance in complex task automation.