Perplexity Reports GPT-6 Astra WANDR Test Results
Company shares benchmark score and cost comparisons from its model evaluation.
TLDR
Perplexity AI posted its evaluation of GPT-6 Astra on WANDR. The model scored 0.682 at $11.98 per task, the highest of any model tested according to the company. Perplexity states that GPT-6 Astra scored 13.5 percent higher than Fable 5.1 at 6.1 percent lower cost. It also scored 27.0 percent higher than Opus 5 at 3.3 percent higher cost. The announcement includes an attached line chart titled GPT-6 Astra on WANDR showing the results.
Combined views
563
4 Sources, first seen ago
