According to the Geekbench listing, the GPU is equipped with 28 Compute Units (CUs), likely representing Work Group Processors (WGPs) in the RDNA 4 architecture. This configuration suggests a total of 56 Compute Units, positioning the Navi 48 between the existing RX 7700 XT (54 CU) and RX 7800 XT (60 CU) models. The clock speed for this sample is listed at 2.1 GHz, which is lower than the current RDNA 3 GPUs, which typically boost to 2.5-2.6 GHz. This lower frequency is likely due to the early stage of testing, with higher clock speeds anticipated closer to the official launch.
The memory specifications include 16 GB of VRAM, consistent with high-end models in the current lineup, likely paired with a 256-bit bus interface. Some variants may feature 12 GB of VRAM with a 192-bit bus. While not yet confirmed, it is expected that AMD will continue to use GDDR6 memory for these GPUs. These early benchmarks offer a glimpse into the potential performance of AMD's RDNA 4 GPUs as the company prepares for their official release next year.
Source: Wccftech