Part 2/8:
Generative AI models that create everything from images to voice and video are increasingly becoming efficient enough to run directly on user devices such as smartphones and tablets. The notion that computing power today can rival that of computers from 10 to 20 years ago underlines the potential of smaller AI models, which can now harness the robust processing capabilities of these devices. These models, featuring from 2 billion to 8 billion parameters, can generate impressive user experiences in real-time.