AMD MI350P
vs
Nvidia RTX 5090
The MI350P is newer, with 4.5x the memory and 2.2x the bandwidth.
At a glance
Memory
4.5x more on MI350P
- MI350P
- 144 GB
- RTX 5090
- 32 GB
Memory bandwidth
2.2x higher on MI350P
- MI350P
- 4.00 TB/s
- RTX 5090
- 1.79 TB/s
Cheapest on-demand
- MI350P
- --
- RTX 5090
- $0.35 /GPU/hr
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | MI350P | RTX 5090 | Difference |
|---|---|---|---|
| On-demand | No on-demand listings | $0.72 /GPU/hr from $0.35 | |
| Reserved | No reserved listings | $0.59 /GPU/hr from $0.21 (1 mo) | |
| Spot | No spot listings | $0.64 /GPU/hr from $0.25 |
Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Medians are across providers, one vote each, so a large catalog does not outweigh a small one. Verify before provisioning. More on how we price.
Price history
MI350P vs RTX 5090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | MI350P | RTX 5090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 144 GB HBM3e | 32 GB GDDR7 | 4.5x more on MI350P |
| Bandwidth | 4,000 GB/s | 1,792 GB/s | 2.2x higher on MI350P |
| Bus width | 4,096-bit | 512-bit | 8x wider on MI350P |
| Compute · dense | |||
| FP4 | 4,600 TFLOPS | 1,676 TFLOPS | 2.7x higher on MI350P |
| FP8 | 2,300 TFLOPS | 419 TFLOPS | 5.5x higher on MI350P |
| FP16 / BF16 | 1,150 TFLOPS | 209.5 TFLOPS | 5.5x higher on MI350P |
| INT8 | 2,300 TOPS | 838 TOPS | 2.7x higher on MI350P |
| FP32 | 72 TFLOPS | 104.8 TFLOPS | 46% higher on RTX 5090 |
| FP64 | 36 TFLOPS | Not supported | MI350P only |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 600 W | 575 W | Within 5% |
| Released | Q2 2026 | Q1 2025 | 15 months newer on MI350P |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16FP32FP64INT8 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI350P datasheet and RTX 5090 datasheet.
Where to rent them
Providers that rent these GPUs, with each one's cheapest rate. No provider we track offers both right now.
| Provider | MI350P | RTX 5090 |
|---|---|---|
HyperAI
Hong Kong
|
-- | $0.35 /GPU/hr On-demand |
Lium
Distributed
|
-- | $0.49 /GPU/hr On-demand |
GPUhub
Singapore
|
-- | $0.52 /GPU/hr On-demand |
GPU.ai
UAE
|
-- | $0.55 /GPU/hr On-demand |
Salad
Distributed
|
-- | $0.60 /GPU/hr On-demand |
Sesterce
France
|
-- | $0.71 /GPU/hr On-demand |
Runcrate
USA
|
-- | $0.72 /GPU/hr On-demand PCIe |
Theta EdgeCloud
Distributed
|
-- | $0.74 /GPU/hr On-demand |
0 providers list the MI350P and 23 the RTX 5090. Every listing is on the MI350P page and the RTX 5090 page.
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
-
-
Nvidia RTX 4090
24 GB GDDR6X · 44% less bandwidth than the RTX 5090
On-demand from $0.33 /GPU/hr
Our data for AMD MI350P was last updated on Nov. 13, 2026, and for Nvidia RTX 5090 on Sept. 14, 2026.