AMD Radeon VII 16GB vs AMD Radeon RX Vega 64 8GB

The Radeon VII delivers 9% more FP16 throughput than the Radeon RX Vega 64 (28 vs 25 TFLOPS dense). The Radeon VII also carries 8 GB more memory (16 GB vs 8 GB).

Radeon VII: Vega, 2019 Radeon RX Vega 64: Vega, 2017
FP16 dense lead
+9%
Radeon VII: 28 vs 25 TFLOPS
VRAM
16 vs 8 GB
8 GB more for the Radeon VII
Bandwidth
1.0 TB/s vs 484 GB/s
Radeon VII moves data faster
TDP
300 W vs 295 W
Radeon RX Vega 64 draws 5 W less

Specifications

Radeon VII
Radeon RX Vega 64
Architecture
VegaVega
Launch Year
20192017
Form Factor
PCIePCIe
VRAM
16 GB8 GB
Memory Bandwidth
1.0 TB/s484 GB/s
TDP
300 W295 W
Process Node
7nm

Performance (TFLOPS)

Radeon VII
Radeon RX Vega 64
FP64
3.5 TFlops No verified data available
FP32
13.8 TFlops 12.7 TFlops
TF32
No verified data available No verified data available
BF16
No verified data available No verified data available
FP16
27.7 TFlops 25.3 TFlops
FP8
No verified data available No verified data available
FP6
No verified data available No verified data available
FP4
No verified data available No verified data available
INT8
No verified data available No verified data available

FLOPS by Precision

What actually differs

The Radeon VII is the newer part: Vega, launched in 2019, against the Radeon RX Vega 64's Vega from 2017. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.

Dense throughput for the Radeon VII against the Radeon RX Vega 64: FP32 14 vs 13 TFLOPS, FP16 28 vs 25 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 16 GB against 8 GB, fed at 1.0 TB/s versus 484 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 300 W for the Radeon VII and 295 W for the Radeon RX Vega 64. At FP32 that works out to 0.05 against 0.04 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.

Frequently asked questions

Is the Radeon VII faster than the Radeon RX Vega 64?

At FP16 precision the Radeon VII reaches 28 TFLOPS dense against 25 TFLOPS for the Radeon RX Vega 64. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the Radeon VII or the Radeon RX Vega 64?

The Radeon VII carries 16 GB of VRAM versus 8 GB for the Radeon RX Vega 64. Memory bandwidth is 1.0 TB/s for the Radeon VII and 484 GB/s for the Radeon RX Vega 64.

How much power do the Radeon VII and the Radeon RX Vega 64 draw?

The Radeon VII is rated at 300 W TDP and the Radeon RX Vega 64 at 295 W. On FP32 throughput per watt, the Radeon VII is the more efficient part.

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.