AMD MI350P
vs
Nvidia H800
The MI350P is newer, with 80% more memory and 2x the bandwidth.
At a glance
Memory
80% more on MI350P
- MI350P
- 144 GB
- H800
- 80 GB
Memory bandwidth
2x higher on MI350P
- MI350P
- 4.00 TB/s
- H800
- 2.00 TB/s
Cheapest on-demand
- MI350P
- --
- H800
- --
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
Specs comparison
| Spec | MI350P | H800 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 144 GB HBM3e | 80 GB HBM2e | 80% more on MI350P |
| Bandwidth | 4,000 GB/s | 2,000 GB/s | 2x higher on MI350P |
| Bus width | 4,096-bit | 5,120-bit | 25% wider on H800 |
| Compute · dense | |||
| FP4 | 4,600 TFLOPS | Not supported | MI350P only |
| FP8 | 2,300 TFLOPS | 1,513 TFLOPS | 52% higher on MI350P |
| FP16 / BF16 | 1,150 TFLOPS | 756.5 TFLOPS | 52% higher on MI350P |
| INT8 | 2,300 TOPS | 1,513 TOPS | 52% higher on MI350P |
| FP32 | 72 TFLOPS | 51 TFLOPS | 41% higher on MI350P |
| FP64 | 36 TFLOPS | ||
| Platform | |||
| Interconnect | PCIe only | NVLink 400 GB/s | NVLink on H800 only |
| Board power | 600 W | 350 W | 71% higher on MI350P |
| Released | Q2 2026 | Q1 2023 | 3 years newer on MI350P |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16FP32FP64INT8 | FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI350P 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 H800 on Sept. 14, 2026.