AMD MI350P
vs
Nvidia P620
The MI350P is newer, with 72x the memory and 50x the bandwidth.
At a glance
Memory
72x more on MI350P
- MI350P
- 144 GB
- P620
- 2 GB
Memory bandwidth
50x higher on MI350P
- MI350P
- 4,000 GB/s
- P620
- 80 GB/s
Cheapest on-demand
- MI350P
- --
- P620
- --
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
Price history
MI350P vs P620 price history
Weekly median on-demand price per GPU per hour
Aggregating historical prices...
Specs comparison
| Spec | MI350P | P620 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 144 GB HBM3e | 2 GB GDDR5 | 72x more on MI350P |
| Bandwidth | 4,000 GB/s | 80 GB/s | 50x higher on MI350P |
| Bus width | 4,096-bit | 128-bit | 32x wider on MI350P |
| Compute · dense | |||
| FP4 | 4,600 TFLOPS | Not supported | MI350P only |
| FP8 | 2,300 TFLOPS | Not supported | MI350P only |
| FP16 / BF16 | 1,150 TFLOPS | ||
| INT8 | 2,300 TOPS | Not supported | MI350P only |
| FP32 | 72 TFLOPS | 1.3 TFLOPS | 55x higher on MI350P |
| FP64 | 36 TFLOPS | Not supported | MI350P only |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 600 W | 40 W | 15x higher on MI350P |
| Released | Q2 2026 | Q1 2018 | 8 years newer on MI350P |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16FP32FP64INT8 | FP16FP32 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI350P datasheet and P620 datasheet.
Similar GPUs
Other comparisons in the same class.
-
-
Nvidia RTX PRO 6000
96 GB GDDR7 · 60% less bandwidth than the MI350P
On-demand from $0.65 /GPU/hr
-
Our data for AMD MI350P was last updated on Nov. 13, 2026, and for Nvidia P620 on Sept. 14, 2026.