A leading chipmaker unveiled its next-generation processor this week, claiming a significant leap in on-device AI performance while using roughly a third less power than its predecessor.
The company says the new architecture is designed specifically for running large language models and image generation tasks locally, without relying on a constant connection to the cloud.
Early benchmark results shared with select partners show substantial gains in tasks like real-time translation and photo processing, though independent testing has yet to confirm the figures.
Devices built around the new chip are expected to reach store shelves by early next year, with several major manufacturers already confirmed as launch partners.
