Last week, Samsung announced the Exynos 2600 chip as the world’s first 2nm chip, and it is expected to debut with the Galaxy S26 series in the first quarter of 2026.
With the initial reveal, additional details have now emerged that highlight the artificial intelligence capabilities of the new chip.
The Exynos 2600 carries a completely new NPU artificial intelligence processor, and Samsung confirms that it offers 113% higher performance compared to the Exynos 2500. In addition to this jump in power, the processor will receive significant support for artificial intelligence tasks on the device itself without relying on an Internet connection.
Korean AI company NotaAI has revealed that its AI model optimization techniques will be integrated within the Exynos 2600 to allow larger models to run directly on the device.
NotaAI relies on the NetsPresso platform to scale down AI models while maintaining the accuracy of results, enabling Samsung to run broader, more advanced models on mobile devices. Thanks to this collaboration, Samsung engineers will be able to significantly scale up local AI capabilities, thus reducing the constant need to connect to the cloud.
The two companies have previously worked together to improve AI tasks in the Exynos 2500 and Exynos 2400 chipsets, and also co-developed the Exynos AI Studio tool, which provides an integrated working environment for building device-optimized AI models.
