Perplexity CEO Unveils AI Operating System That Coordinates Up to 20 Models

Perplexity CEO Aravind Srinivas has outlined the company's vision for AI operating systems capable of coordinating multiple artificial intelligence models, tools and files while balancing privacy, cost and performance.

Speaking during Intel CEO Lip-Bu Tan's keynote at COMPUTEX 2026 on Tuesday, Srinivas said Perplexity's recently launched Perplexity Computer can create a team of AI agents and use up to 20 AI models within a single workflow.

"It creates a team of agents, uses up to 20 different AI models, and it orchestrates across models, tools and files in one single system," he said.

Srinivas explained that the platform's "agent harness" uses what he called hybrid agentic inference to coordinate models while optimising intelligence, accuracy, privacy and cost. The approach allows smaller AI models to run locally on devices, while more demanding tasks can be handled in data centres.

His remarks reflect a broader industry push toward hybrid AI computing , which combines on-device processing with cloud-based infrastructure to improve efficiency and privacy.

"What we are showing today is just the start," Srinivas said, adding that future AI systems will rely on growing computing power in both local machines and data centres.

Calling the technology a milestone in AI and chip engineering, Srinivas also thanked Intel for its partnership in advancing the project.

Earlier, Tan highlighted the growing role of AI in computing and identified on-device AI as a major focus area for Intel.

"AI is profoundly impacting the way we use our devices. A major focus area for us is the use of AI on devices. Together with partners, we are at the forefront of advancing intelligence," Tan said.