AMD MI250
vs
Nvidia GTX 1660 Ti
The MI250 is newer, with 21x the memory and 11x the bandwidth.
At a glance
Memory
21x more on MI250
- MI250
- 128 GB
- GTX 1660 Ti
- 6 GB
Memory bandwidth
11x higher on MI250
- MI250
- 3,200 GB/s
- GTX 1660 Ti
- 288 GB/s
Cheapest on-demand
- MI250
- --
- GTX 1660 Ti
- --
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
Price history
MI250 vs GTX 1660 Ti price history
Weekly median on-demand price per GPU per hour
Aggregating historical prices...
Specs comparison
| Spec | MI250 | GTX 1660 Ti | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 128 GB HBM2e | 6 GB GDDR6 | 21x more on MI250 |
| Bandwidth | 3,200 GB/s | 288 GB/s | 11x higher on MI250 |
| Bus width | 8,192-bit | 192-bit | 43x wider on MI250 |
| Compute · dense | |||
| FP16 / BF16 | 362.1 TFLOPS | 10.9 TFLOPS | 33x higher on MI250 |
| INT8 | 362.1 TOPS | Not supported | MI250 only |
| FP32 | 90.5 TFLOPS | 5.4 TFLOPS | 17x higher on MI250 |
| FP64 | 45.3 TFLOPS | Not supported | MI250 only |
| Platform | |||
| Interconnect | Infinity Fabric 400 GB/s | PCIe only | Infinity Fabric on MI250 only |
| Board power | 500 W | 120 W | 4.2x higher on MI250 |
| Released | Q4 2021 | Q1 2019 | 3 years newer on MI250 |
| Precision | |||
| Supported formats | FP16BF16FP32FP64INT4INT8 | FP16FP32 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI250 datasheet and GTX 1660 Ti datasheet.
Similar GPUs
Other comparisons in the same class.
-
-
-
-
Nvidia RTX 3060
12 GB GDDR6 · 25% more bandwidth than the GTX 1660 Ti
On-demand from $0.08 /GPU/hr -
Nvidia RTX 2060
6 GB GDDR6 · 17% more bandwidth than the GTX 1660 Ti
On-demand from $0.06 /GPU/hr
Our data for AMD MI250 was last updated on Sept. 14, 2026, and for Nvidia GTX 1660 Ti on Nov. 13, 2026.