US National WireUS NATIONAL WIRE
Tech

Perplexity Expands Model Council to Cloud-Based Computer Platform

Portrait of Simone Larkin
Simone Larkinthe futuristJul 28AI
Perplexity Expands Model Council to Cloud-Based Computer Platform

AI-generated image · US National Wire

The AI search firm now allows users to assemble a 'board' of up to eight competing models to synthesize answers for ambiguous business problems.

Perplexity has expanded its Model Council feature to its cloud-based Computer platform, enabling users to configure multi-model queries to tackle complex business issues. According to reporting from The Register, the tool allows users to build an AI advisory board consisting of two to eight models. These can include open-weight models such as Kimi and GLM, as well as frontier lab models from Google, Anthropic, and OpenAI.

Perplexity communications chief Jesse Dwyer explained to The Register that the system functions by using an orchestrator model to deploy parallel subagents. Each agent tackles a prompt independently before a user-selected "chair" model synthesizes the outputs. This process explicitly highlights areas of consensus and divergence, which Perplexity states is ideal for situations involving risk assessment, tradeoffs, or judgment calls where no single correct answer exists.

Integrated into the Computer interface, the Model Council can transform these syntheses into work-ready assets, including board decks and reports. Perplexity suggests the tool for use in engineering, financial queries, corporate decision-making, and legal questions. Users can further refine results by asking for confidence levels or requesting that models state the underlying assumptions behind their conclusions.

Access is available to Pro ($20/month) and Max ($200/month) individual users, as well as enterprise tiers. Perplexity's Enterprise Pro costs $34 per seat monthly, while Enterprise Max is $271. The system utilizes usage-based billing via Computer credits, where 100 credits equal $1. While simple queries may cost less than a penny, "heavy" tasks—such as creating pitch decks—can consume up to 2,275 credits, according to The Register.

Sources

More from Simone Larkin