Flopper.io GPU Comparison Sheet
NVIDIA RTX PRO 6000 Blackwell Server Edition vs NVIDIA GeForce RTX 5090 32GB
Side by side datacenter GPU specifications from the Flopper.io database.
Overview
| NVIDIA RTX PRO 6000 Blackwell Server Edition | NVIDIA GeForce RTX 5090 32GB | |
|---|---|---|
| FP32 TFLOPS | 120.0 | 104.8 |
| VRAM | 96 GB | 32 GB |
| Memory Bandwidth | 1.6 TB/s | 1.8 TB/s |
| TDP | 600 W | 575 W |
Performance by Precision
| Precision | NVIDIA RTX PRO 6000 Blackwell Server Edition peak TFLOPS (dense) | NVIDIA GeForce RTX 5090 32GB peak TFLOPS (dense) |
|---|---|---|
| FP64 | — | 1.6 |
| FP32 | 120.0 | 104.8 |
| TF32 | 234.0 | 104.8 |
| BF16 | — | 209.5 |
| FP16 | — | 419.1 |
| FP8 | — | 838.2 |
| FP6 | — | — |
| FP4 | — | 1676.0 |
| INT8 | — | 838.2 |
Specifications
| NVIDIA RTX PRO 6000 Blackwell Server Edition | NVIDIA GeForce RTX 5090 32GB | |
|---|---|---|
| Architecture | Blackwell | Blackwell |
| Process Node | 4nm | 4nm |
| Launch Year | 2025 | 2025 |
| Form Factor | PCIe | PCIe |
| Memory Type | GDDR7 | GDDR7 |
| Interface Width | 512-bit | 512-bit |
| Interconnect | — | — |
| TDP | 600 W | 575 W |
| FP32 Efficiency | 0.20 TFLOPS/W | 0.18 TFLOPS/W |