Perplexity has unveiled a significant upcoming feature for its AI system, Perplexity Computer, which will allow users to split tasks between local and cloud-based AI models. This innovative hybrid approach aims to combine the security of local processing with the power of server-based AI for various tasks. The feature is particularly beneficial for handling sensitive information, such as financial records and health data, by ensuring that such data is processed locally when necessary. The compact model running on the device will intelligently determine when to keep data local, while more complex tasks will leverage the capabilities of server-based models. Unlike traditional systems that require users to make a choice between local or cloud processing beforehand, Perplexity Computer will manage this automatically, optimizing performance based on the specific requirements of each task. This hybrid AI orchestration feature is anticipated to be rolled out in July, marking a significant advancement in AI task management.