Perplexity Computer adds multiagent mode for complex tasks

Executive Briefing
“Perplexity has introduced a new operating mode allowing up to 16 AI agents to coordinate...”
Key Takeaways
- 1.The Multiagent mode allows a lead agent to coordinate up to 16 supporting agents working in parallel.
- 2.The default team combines models from different developers such as Claude Opus 5.5, GPT 6.1 Sol, and Gemini 4 Carbon.
- 3.Agents can share findings, question assumptions, and collaboratively edit shared data files.
Perplexity has introduced a new operating mode allowing up to 16 AI agents to coordinate directly. The feature is currently available exclusively to Max plan users on the web platform.
Key Developments & Primary Facts
The Multiagent mode allows a lead agent to coordinate up to 16 supporting agents working in parallel.
The default team combines models from different developers such as Claude Opus 5.5, GPT 6.1 Sol, and Gemini 4 Carbon.
Agents can share findings, question assumptions, and collaboratively edit shared data files.
Strategic Analysis & Implications
Combining multiple foundational models in a single team leverages the distinct strengths of each system rather than depending on a single vendor.
Direct communication among agents establishes a closed-loop process for self-checking errors before synthesizing results for the user.
Outlook & Factors to Watch
The company has not published benchmark metrics regarding the accuracy or processing speed of this mode.
The specific credit consumption cost per multiagent task has not been disclosed.







