Perplexity said Multiagent is available now to Max and Enterprise Max subscribers on the web, with mobile and desktop support coming soon. The company calls the feature credit-intensive and recommends it only for important tasks. On Terminal-Bench 4.1, the mixed-model team led by Gemini 4 Carbon scored 0.727, the highest result. Every mixed team beat four copies of its lead model, by +0.008 to +0.125 on Terminal-Bench 4.1 and +0.015 to +0.064 on WANDR. The top mixed team on each benchmark also cost less than four copies of its lead: $10.31 versus roughly $13 per run, and $54.50 versus $66.41. Perplexity said three of four teams tested on WANDR finished sooner than the model alone.

AI summary of the linked source — not editorial content.

Explore the topic

Loading related coverage…

Comments

Loading comments…