Flopper.io GPU Comparison Sheet
NVIDIA GeForce RTX 4070 SUPER 12GB vs NVIDIA RTX 6000 Ada 48GB
Side by side datacenter GPU specifications from the Flopper.io database.
Overview
| NVIDIA GeForce RTX 4070 SUPER 12GB | NVIDIA RTX 6000 Ada 48GB | |
|---|---|---|
| FP32 TFLOPS | 35.5 | 91.1 |
| VRAM | 12 GB | 48 GB |
| Memory Bandwidth | — | 960 GB/s |
| TDP | 220 W | 300 W |
Performance by Precision
| Precision | NVIDIA GeForce RTX 4070 SUPER 12GB peak TFLOPS (dense) | NVIDIA RTX 6000 Ada 48GB peak TFLOPS (dense) |
|---|---|---|
| FP64 | 0.6 | — |
| FP32 | 35.5 | 91.1 |
| TF32 | 35.5 | — |
| BF16 | 71.0 | — |
| FP16 | 71.0 | — |
| FP8 | 141.9 | 728.5 |
| FP6 | — | — |
| FP4 | — | — |
| INT8 | 283.9 | — |
Specifications
| NVIDIA GeForce RTX 4070 SUPER 12GB | NVIDIA RTX 6000 Ada 48GB | |
|---|---|---|
| Architecture | Ada Lovelace | Ada Lovelace |
| Process Node | 4nm | 4nm |
| Launch Year | 2024 | 2023 |
| Form Factor | PCIe | PCIe |
| Memory Type | GDDR6X | — |
| Interface Width | 192-bit | — |
| Interconnect | — | — |
| TDP | 220 W | 300 W |
| FP32 Efficiency | 0.16 TFLOPS/W | 0.30 TFLOPS/W |