Perplexity Unveils “Perplexity Computer,” a System to Orchestrate Multiple AI Models

February 25, 2026 - AI research and development company Perplexity has announced Perplexity Computer, a new system that integrates multiple leading AI models into a single workflow. The system acts as a general-purpose digital worker capable of autonomously creating and executing complex tasks that can run for hours or even months.

A Digital Worker That Coordinates AI Models

Perplexity Computer operates software stacks similarly to a human worker. Users describe a desired outcome, and the system breaks it down into tasks and subtasks, assigning them to sub-agents. These sub-agents handle web research, document generation, data processing, or API interactions. Work is coordinated automatically and runs asynchronously.

If a problem arises, the system generates additional sub-agents to resolve it and only checks in with the user if necessary. Tasks run in isolated environments with access to real file systems, browsers, and tool integrations, ensuring safe execution.

Intelligent Multi-Model Orchestration

The system does not rely on a single AI model. Instead, it selects and orchestrates the best models for each task. The current setup includes:

• Opus 4.6: Core reasoning engine
• Gemini: Deep research and sub-agent creation
• Nano Banana: Image generation
• Veo 3.1: Video generation
• Grok: High-speed processing for lightweight tasks
• ChatGPT 5.2: Long-context memory and wide-range search

Users can also specify which models to use for specific subtasks. The system is model-agnostic, allowing flexibility as AI models continue to evolve.

Availability

Perplexity Computer is currently available to Perplexity Max subscribers, with plans to roll it out to Enterprise Max users in the near future.