Perplexity Introduces Multiagent System to Cross-Check AI Work
Perplexity has launched Multiagent for its Computer platform, allowing up to 16 AI agents to collaborate, share findings, and cross-examine each other's work across different frontier models.
Key points
- Deploys up to 16 parallel AI agents coordinated by a single lead model
- Features direct peer-to-peer communication and shared file editing across Claude, GPT, and Gemini models
- Available exclusively on the web for Perplexity Max subscribers
Search and research platform Perplexity has rolled out Multiagent, a new highest-tier effort level for its Computer platform designed to handle complex, multistep research and analysis. Positioned above the existing Ultra setting, the feature enables a lead orchestrator model to manage a collaborative team of up to 16 distinct AI agents operating simultaneously in the cloud.
Unlike previous configurations where supporting agents only reported back to a single leader, this new architecture allows agents to communicate directly with one another. They can share evidence, question underlying assumptions, flag data discrepancies, and coordinate live edits on shared files—such as updating financial revenue figures that another agent relies upon for a business valuation. The default team utilizes various frontier families, including Claude, GPT, and Gemini models, while users maintain a single conversational point of contact with the lead orchestrator.
The platform is currently accessible exclusively on the web for Perplexity Max subscribers, with no local or quantized execution options provided. Perplexity has not released formal benchmark data regarding speed or accuracy metrics for the multiagent setup.
What to watch next: Monitor user adoption and real-world performance reports regarding how effectively these multi-model teams handle complex enterprise data room analyses.
Source MarkTechPost — marktechpost.com/2026/11/17/perplexity-computer-multiagent-effort-l… ↗
Read next
🤖AI ModelsGoogle Expands Interactive World-Building with Genie 4 Launch
Google has released its latest world-generation model, Genie 4, to premium AI Ultra subscribers, introducing real-time environment modification and audio integration.
🏢Big TechMicrosoft Challenges Rivals on AI Pricing at Ignite Conference
Microsoft is positioning its Copilot AI as the more cost-effective choice for businesses compared to OpenAI and Anthropic, citing a commissioned study to support its clai…
🚀Startups & FundingOpenAI CEO Sam Altman Addresses Upcoming Biographical Film
OpenAI CEO Sam Altman commented on the upcoming film 'Artificial,' which portrays his life and the 2023 leadership crisis at the company.
🌐Work & SocietyDesign Studio Office Commun Launches Minimalist WebKit Browser for macOS
Paris-based design studio Office Commun has released Search, a lightweight, open-source web browser for macOS that prioritizes performance and minimalist design.
🔌Chips & HardwareSnapchat Integrates AI Coding Assistants into AR Lens Development
Snapchat has launched the Easy Lens CLI, enabling developers to build augmented reality experiences using AI coding tools like Claude Code and Cursor.
🚀Startups & FundingRadixArk Unveils DaisyRose-27B, Focused on Agentic Efficiency
Startup RadixArk has launched DaisyRose-27B, an AI model optimized for long-horizon agent tasks, claiming significant improvements in training efficiency and reasoning ca…