AI search startup Perplexity is expanding beyond traditional question-and-answer experiences with the launch of a new AI agent called Computer, built specifically for high-value enterprise work.
Instead of relying on a single large model to handle every task, Computer orchestrates 19 different AI models and coordinates them as a team. The idea is simple: different models are better at different things, so a multi-model approach can deliver more reliable results across complex business scenarios. Perplexity says this system is designed to take advantage of that specialization—routing each part of a job to the model best suited for it, then combining the outputs into a more complete response.
For organizations looking to put AI to work in real operations, the pitch is clear: enterprise use cases often require more than just fluent text generation. They demand accuracy, consistency, control, and the ability to handle varied tasks that may include research-style querying, structured reasoning, summarization, and other specialized capabilities. By bundling 19 models into one coordinated agent, Perplexity aims to improve performance on these higher-stakes workloads where a single general-purpose model may fall short.
The move also reflects a broader shift in the AI industry: businesses are increasingly evaluating AI systems not just by raw model size, but by how well they can complete multi-step tasks, adapt to different problem types, and deliver dependable outcomes. A coordinated agent that can pick the right tool for each step is positioned as a practical approach for enterprises that care about results and repeatability.
With Computer, Perplexity is signaling that the future of AI at work may look less like one “all-purpose brain” and more like a managed system of specialized models—assembled, directed, and optimized to handle demanding real-world tasks at scale.
