OpenAI's GPT-5.6 Sol and Luna outperform Terra at every point on the Intelligence vs Cost per Task chart, with Luna noted as "particularly cost efficient." All three GPT-5.6 models push past GPT-5.5 on the Pareto frontier. Separately, GPT-5.6 Sol ranks second to Claude Fable 5 on the Artificial Analysis Intelligence Index at one third of the cost, while Meta's Muse Spark 1.1 scores 51 on the same index.
No score is assigned. Sources and their independence are shown in the citation chain below.