Ploy Makes GPT-5.6 Sol Its Default Model After 2.2× Faster Production Tests

Ploy has made OpenAI’s GPT-5.6 Sol the default model for every Ploy workspace after a head-to-head production evaluation found its marketing-site agent completed builds in 3 minutes 42 seconds, versus 8 minutes on Claude Opus 4.8. The company reported 27% lower cost and a higher visual score, but only after reworking model-specific tooling and evaluation assumptions.