Flopper.io GPU Comparison Sheet
NVIDIA GB200 Grace Blackwell Superchip 372GB vs NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Side by side datacenter GPU specifications from the Flopper.io database.
Overview
| NVIDIA GB200 Grace Blackwell Superchip 372GB | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | |
|---|---|---|
| FP32 TFLOPS | 160.0 | 109.7 |
| Memory | 372 GB | 96 GB |
| Memory Bandwidth | 16.0 TB/s | 1.8 TB/s |
| TDP | — | 300 W |
Performance by Precision
| Precision | NVIDIA GB200 Grace Blackwell Superchip 372GB peak TFLOPS (dense) | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition peak TFLOPS (dense) |
|---|---|---|
| FP64 | 80.0 | 1.7 |
| FP32 | 160.0 | 109.7 |
| TF32 | 2500.0 | 219.5 |
| BF16 | 5000.0 | 438.9 |
| FP16 | 5000.0 | 438.9 |
| FP8 | 10000.0 | 877.9 |
| FP6 | 10000.0 | — |
| FP4 | 20000.0 | 1755.7 |
| INT8 | 10000.0 | 877.9 |
Specifications
| NVIDIA GB200 Grace Blackwell Superchip 372GB | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | |
|---|---|---|
| Architecture | Blackwell | Blackwell |
| Process Node | 4nm | 4nm |
| Launch Year | 2024 | 2025 |
| Form Factor | Superchip | PCIe |
| Memory Type | HBM3e | GDDR7 |
| Interface Width | — | 512-bit |
| Interconnect | NVLink-C2C | — |
| TDP | — | 300 W |
| FP32 Efficiency | — | 0.37 TFLOPS/W |