Samsung is poised to unveil its next-generation Exynos 2600 system-on-chip (SoC) alongside the Galaxy S26 on . The chip is already generating significant attention, with early indications suggesting it could rival, and in some cases surpass, Qualcomm’s Snapdragon 8 Elite Gen 5 in performance. A key factor driving this potential leap forward is the integration of Arm’s Scalable Matrix Extensions 2 (SME2) technology.
Arm has officially confirmed that the Exynos 2600 will support SME2, a development that provides insight into the chip’s underlying architecture and capabilities. According to Arm, SME2 will accelerate AI-powered applications, such as object detection, by up to 70%. This enhancement is particularly significant as on-device AI processing becomes increasingly central to the mobile experience.
Arm introduced SME2 alongside its new Lumex CPU cores – the C1-Ultra, C1-Premium, C1-Pro, and C1-Nano. The Exynos 2600 is expected to utilize a combination of these cores. Multiple Geekbench listings have indicated a 10-core processor configuration. Based on previous Exynos chip designs, this is likely to consist of one Lumex C1-Ultra core, three Lumex C1-Premium cores, and six Lumex C1-Nano cores.
Stefan Rosinger of Arm elaborated on the benefits of this configuration, stating, “As on-device AI becomes a central element of the mobile experience, efficiency and responsiveness are increasingly crucial. Built on Arm compute subsystems with SME2-enabled C1-Ultra and C1-Pro, Exynos 2600 leverages SME2 to extend the potential of CPU-based AI, reducing the latency associated with offloading to discrete accelerators and making it well-suited for short, interactive, and real-time AI workloads.”
While the statement focuses specifically on the C1-Ultra and C1-Pro cores, it implicitly suggests that the Exynos 2600 might not utilize a full complement of Lumex C1-Premium cores. This could be a strategic decision by Samsung to manage thermal constraints, as incorporating more premium cores could increase the chip’s heat output.
The success of the Exynos 2600 is critical for Samsung Foundry’s SF2 node and its ambition to attract major customers like Qualcomm and Nvidia. Samsung Foundry recently secured a deal with Tesla for automotive chips, demonstrating its growing capabilities in advanced semiconductor manufacturing. The 2nm GAA process represents a significant step forward offering potential improvements in performance, power efficiency, and area reduction compared to the previous 3nm GAA process.
Samsung’s 2nm GAA process is aiming to be a competitive alternative to TSMC’s 2nm technology. Recent reports indicate that Samsung has made substantial progress in improving yields on its 2nm node, reaching 50 percent after initial yields of 30 percent. This improvement enabled the company to begin mass production of the Exynos 2600 in late September, suggesting further yield improvements are possible.
The 2nm GAA process is projected to deliver up to a 12 percent performance increase, a 25 percent improvement in power efficiency, and a 5 percent reduction in chip area compared to the 3nm GAA process. These advancements are crucial for enabling more powerful and efficient mobile devices, as well as supporting increasingly complex AI workloads.
The Galaxy S26 series is expected to follow a familiar three-model structure. While details regarding specific model configurations remain fluid, the Galaxy S26 Ultra appears to be the most stable in terms of design. The Exynos 2600 will likely be featured in select Galaxy S26 models, alongside Qualcomm’s Snapdragon 8 Elite Gen 5 in other regions, continuing Samsung’s strategy of utilizing both in-house and third-party processors.
Samsung is also offering a pre-registration incentive for the Galaxy S26 series, providing a voucher worth approximately $32 USD (equivalent to RM150) for future purchases of the new devices and other select Galaxy products. Pre-registration closes on .
