A consumer GPU against a workstation card, one Kepler and one Maxwell: both with 2 GB DDR3 and 2.1x the memory bandwidth on the K620.
At a glance
Memory
- GT 730
- 2 GB
- K620
- 2 GB
Identical
Memory bandwidth
- GT 730
- 14 GB/s
- K620
- 29 GB/s
2.1x higher on K620
Cheapest on-demand
- GT 730
- --
- K620
- --
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
Price history
GT 730 vs K620 price history
Weekly median on-demand price per GPU per hour
Aggregating historical prices...
Specs comparison
| Spec | GT 730 | K620 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 2 GB DDR3 | 2 GB DDR3 | Identical |
| Bandwidth | 14 GB/s | 29 GB/s | 2.1x higher on K620 |
| Bus width | 64-bit | 128-bit | 2x wider on K620 |
| Compute · dense | |||
| FP32 | 0.7 TFLOPS | 0.9 TFLOPS | 1.2x higher on K620 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 23 W | 45 W | 2x higher on K620 |
| Released | Q2 2014 | Q3 2014 | |
| Precision | |||
| Supported formats | FP32 | FP32 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: K620 datasheet.
Similar GPUs
Other comparisons in the same class.
-
Nvidia GTX 1050 Ti
From $0.02 /GPU/hr Compare
Our data for Nvidia GT 730 was last updated on Sept. 7, 2026, and for Nvidia K620 on Sept. 7, 2026.