Perplexity Computer adding ability to split tasks between local and cloud models - 9to5Mac

Perplexity splits AI inference between PCs and cloud to cut costs

Perplexity AI launches a hybrid platform routing AI inference between PCs and cloud in real time to slash costs amid the $500M/month compute crisis. The approach cuts marginal cost per query by offloading lightweight tasks to edge, a competitive variable as inference demand strains infrastructure.