AMD Radeon PRO W7900 48GB vs AMD Radeon RX 7900 GRE 16GB

Both GPUs launched in 2023: the Radeon PRO W7900 on RDNA 3 and the Radeon RX 7900 GRE on RDNA 3.

Dense throughput for the Radeon PRO W7900 against the Radeon RX 7900 GRE: FP32 61 vs 46 TFLOPS, FP16 123 vs 92 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 48 GB against 16 GB, fed at 864 GB/s versus 576 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.

Power budgets are 295 W for the Radeon PRO W7900 and 260 W for the Radeon RX 7900 GRE. At FP32 that works out to 0.21 against 0.18 TFLOPS per watt.

For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.