Samsung will continue to push AI functionality next year, involving the Exynos 2500, according to the X poster Connor/コナー. Samsung will also use a Google-designed TPU, just as it uses AMD-designed GPUs for its flagship chips.
It is incorporated into the Tensor chipsets that are co-developed with Samsung and has been part of Pixel phones for several years. If you are working with artificial intelligence, you can even rent bigger versions of the TPU from Google Cloud. The TPU is one of several types of AI accelerators and is heavily used by Google products.
Using the Exynos 2500 chipset, phones such as the Galaxy S25 series will be able to access Google’s machine-learning APIs. Even though Samsung beefed up the AI hardware on the Exynos 2400, it’s still not enough to power Galaxy AI.
In addition to two Neural Processing Units – G-NPU and S-NPU – the new flagship chipset is expected to include. As the name implies, the “G” stands for “general” and is a more flexible version of a fixed-function NPU. Since 2200, they have been part of Exynos.
According to a recent report, the Galaxy S25 series will use both Exynos 2500 and Snapdragon 8 Gen 4 chips, not just Exynos. In addition to the TPU, Qualcomm has also been pushing AI acceleration – the 8 Gen 3 can run models with up to 10 billion parameters.
Specs:
It has been suggested that Samsung’s second generation 3nm node (3nm GAP/SF3) will be used to fabricate the Exynos 2500. There are four Cortex-X5 cores at 3.2GHz, three A730 cores at 2.3-2.5GHz, and two more A730 cores at a lower clock speed. No details are available on the GPU, but it will use an Xclipse 950.
Comparatively, Samsung’s 4LPP+ node will be used to fabricate the Tensor G4 (which will likely be the source of the TPU design). So the Exynos 2500 should be faster and more efficient, even in AI tasks (which was Google’s forte until now).
