NVIDIA GeForce RTX 5070 Ti 16GB vs NVIDIA GeForce RTX 5090 32GB

The GeForce RTX 5090 delivers 2.4x the FP8 throughput of the GeForce RTX 5070 Ti (838 vs 352 TFLOPS dense). The GeForce RTX 5090 also carries 16 GB more memory (32 GB vs 16 GB).

GeForce RTX 5070 Ti: Blackwell, 2025 GeForce RTX 5090: Blackwell, 2025
FP8 dense lead
2.4x
GeForce RTX 5090: 838 vs 352 TFLOPS
VRAM
16 vs 32 GB
16 GB more for the GeForce RTX 5090
Bandwidth
896 GB/s vs 1.8 TB/s
GeForce RTX 5090 moves data faster
TDP
300 W vs 575 W
GeForce RTX 5070 Ti draws 275 W less

Specifications

GeForce RTX 5070 Ti
GeForce RTX 5090
Architecture
BlackwellBlackwell
Launch Year
20252025
Form Factor
PCIePCIe
VRAM
16 GB32 GB
Memory Bandwidth
896 GB/s1.8 TB/s
TDP
300 W575 W
Process Node
4nm4nm

Performance (TFLOPS)

GeForce RTX 5070 Ti
GeForce RTX 5090
FP64
0.7 TFlops 1.6 TFlops
FP32
43.9 TFlops 104.8 TFlops
TF32
43.9 TFlops
87.9 TFLOPS sparse
104.8 TFlops
209.5 TFLOPS sparse
BF16
87.9 TFlops
175.8 TFLOPS sparse
209.5 TFlops
419.0 TFLOPS sparse
FP16
175.8 TFlops
351.5 TFLOPS sparse
419.1 TFlops
sparse not published
FP8
351.5 TFlops
703.0 TFLOPS sparse
838.2 TFlops
sparse not published
FP6
No verified data available No verified data available
FP4
703.0 TFlops
1406.0 TFLOPS sparse
1676.0 TFlops
sparse not published
INT8
351.5 TFlops
703.0 TOPS sparse
838.2 TFlops
sparse not published

FLOPS by Precision

What actually differs

Both GPUs launched in 2025: the GeForce RTX 5070 Ti on Blackwell and the GeForce RTX 5090 on Blackwell.

Dense throughput for the GeForce RTX 5070 Ti against the GeForce RTX 5090: FP64 0.7 vs 1.6 TFLOPS, FP32 44 vs 105 TFLOPS, FP16 176 vs 419 TFLOPS, FP8 352 vs 838 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 16 GB against 32 GB, fed at 896 GB/s versus 1.8 TB/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 GeForce RTX 5070 Ti and 575 W for the GeForce RTX 5090. At FP32 that works out to 0.15 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.

Frequently asked questions

Is the GeForce RTX 5070 Ti faster than the GeForce RTX 5090?

At FP8 precision the GeForce RTX 5090 reaches 838 TFLOPS dense against 352 TFLOPS for the GeForce RTX 5070 Ti. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the GeForce RTX 5070 Ti or the GeForce RTX 5090?

The GeForce RTX 5090 carries 32 GB of VRAM versus 16 GB for the GeForce RTX 5070 Ti. Memory bandwidth is 896 GB/s for the GeForce RTX 5070 Ti and 1.8 TB/s for the GeForce RTX 5090.

How much power do the GeForce RTX 5070 Ti and the GeForce RTX 5090 draw?

The GeForce RTX 5070 Ti is rated at 300 W TDP and the GeForce RTX 5090 at 575 W. On FP32 throughput per watt, the GeForce RTX 5090 is the more efficient part.

Can I rent the GeForce RTX 5070 Ti or the GeForce RTX 5090 in the cloud?

Yes. Live cloud listings tracked by Flopper start at $0.11 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA GeForce RTX 5070 Ti 16GB

ProviderConfigurationPrice/GPU-hrChecked
Vast.ai
4× GeForce RTX 5070 Ti 16GB Community $0.11 3h ago View →
All GeForce RTX 5070 Ti listings and price history →

NVIDIA GeForce RTX 5090 32GB

ProviderConfigurationPrice/GPU-hrChecked
Vast.ai
2× GeForce RTX 5090 32GB Community $0.33 3h ago View →
Excess Supply
1× GeForce RTX 5090 32GB $0.65 3h ago View →
RunPod
1× GeForce RTX 5090 32GB Community $0.69 3h ago View →
Phyntec
8× GeForce RTX 5090 32GB $0.70 3h ago View →
RunPod
1× GeForce RTX 5090 32GB $0.99 3h ago View →
All GeForce RTX 5090 listings and price history →

Get Comparison Updates

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