NVIDIA GeForce RTX 4070 12GB vs NVIDIA GeForce RTX 4090 D 24GB

The GeForce RTX 4090 D delivers 2.5x the FP8 throughput of the GeForce RTX 4070 (294 vs 117 TFLOPS dense). The GeForce RTX 4090 D also carries 12 GB more memory (24 GB vs 12 GB).

GeForce RTX 4070: Ada Lovelace, 2023 GeForce RTX 4090 D: Ada Lovelace, 2023
FP8 dense lead
2.5x
GeForce RTX 4090 D: 294 vs 117 TFLOPS
VRAM
12 vs 24 GB
12 GB more for the GeForce RTX 4090 D
TDP
200 W vs 425 W
GeForce RTX 4070 draws 225 W less

Specifications

GeForce RTX 4070
GeForce RTX 4090 D
Architecture
Ada LovelaceAda Lovelace
Launch Year
20232023
Form Factor
PCIePCIe
VRAM
12 GB24 GB
Memory Bandwidth
504 GB/s
TDP
200 W425 W
Process Node
4nm4nm

Performance (TFLOPS)

GeForce RTX 4070
GeForce RTX 4090 D
FP64
0.5 TFlops 1.1 TFlops
FP32
29.1 TFlops 73.5 TFlops
TF32
29.1 TFlops
58.2 TFLOPS sparse
73.5 TFlops
sparse not published
BF16
58.3 TFlops
116.6 TFLOPS sparse
147.1 TFlops
sparse not published
FP16
58.3 TFlops
116.6 TFLOPS sparse
147.1 TFlops
sparse not published
FP8
116.6 TFlops
233.2 TFLOPS sparse
294.2 TFlops
sparse not published
FP6
No verified data available No verified data available
FP4
No verified data available No verified data available
INT8
233.2 TFlops
466.4 TOPS sparse
588.4 TFlops
sparse not published

FLOPS by Precision

What actually differs

Both GPUs launched in 2023: the GeForce RTX 4070 on Ada Lovelace and the GeForce RTX 4090 D on Ada Lovelace.

Dense throughput for the GeForce RTX 4070 against the GeForce RTX 4090 D: FP64 0.5 vs 1.1 TFLOPS, FP32 29 vs 74 TFLOPS, FP16 58 vs 147 TFLOPS, FP8 117 vs 294 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 12 GB against 24 GB. 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 200 W for the GeForce RTX 4070 and 425 W for the GeForce RTX 4090 D. At FP32 that works out to 0.15 against 0.17 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 GeForce RTX 4070 faster than the GeForce RTX 4090 D?

At FP8 precision the GeForce RTX 4090 D reaches 294 TFLOPS dense against 117 TFLOPS for the GeForce RTX 4070. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the GeForce RTX 4070 or the GeForce RTX 4090 D?

The GeForce RTX 4090 D carries 24 GB of VRAM versus 12 GB for the GeForce RTX 4070.

How much power do the GeForce RTX 4070 and the GeForce RTX 4090 D draw?

The GeForce RTX 4070 is rated at 200 W TDP and the GeForce RTX 4090 D at 425 W. On FP32 throughput per watt, the GeForce RTX 4090 D is the more efficient part.

Can I rent the GeForce RTX 4070 or the GeForce RTX 4090 D in the cloud?

Yes. Live cloud listings tracked by Flopper start at $0.07 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA GeForce RTX 4070 12GB

ProviderConfigurationPrice/GPU-hrChecked
Vast.ai
1× GeForce RTX 4070 12GB Community $0.07 1h ago View →
All GeForce RTX 4070 listings and price history →

NVIDIA GeForce RTX 4090 D 24GB

ProviderConfigurationPrice/GPU-hrChecked
Vast.ai
1× GeForce RTX 4090 D 24GB Community $0.40 1h ago View →
All GeForce RTX 4090 D listings and price history →

Get Comparison Updates

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