AMD MI250
vs
Nvidia RTX 3090
The MI250 is newer, with 5.3x the memory and 3.4x the bandwidth.
At a glance
Memory
5.3x more on MI250
- MI250
- 128 GB
- RTX 3090
- 24 GB
Memory bandwidth
3.4x higher on MI250
- MI250
- 3,200 GB/s
- RTX 3090
- 936 GB/s
Cheapest on-demand
MI250 is not sold on demand
- MI250
- --
- RTX 3090
- $0.10 /GPU/hr
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | MI250 | RTX 3090 | Difference |
|---|---|---|---|
| On-demand | No on-demand listings | $0.19 /GPU/hr from $0.10 | |
| Reserved | -- /GPU/hr from $1.30 (12 mo) | -- /GPU/hr from $0.06 (1 mo) | |
| Spot | No spot listings | -- /GPU/hr from $0.05 |
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. No median for reserved on the MI250 and reserved and spot on the RTX 3090: we only show one when at least 3 providers list the GPU on that billing type. Verify before provisioning. More on how we price.
Price history
MI250 vs RTX 3090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | MI250 | RTX 3090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 128 GB HBM2e | 24 GB GDDR6X | 5.3x more on MI250 |
| Bandwidth | 3,200 GB/s | 936 GB/s | 3.4x higher on MI250 |
| Bus width | 8,192-bit | 384-bit | 21x wider on MI250 |
| Compute · dense | |||
| FP16 / BF16 | 362.1 TFLOPS | 71.2 TFLOPS | 5.1x higher on MI250 |
| INT8 | 362.1 TOPS | 284.7 TOPS | 27% higher on MI250 |
| FP32 | 90.5 TFLOPS | 35.6 TFLOPS | 2.5x higher on MI250 |
| FP64 | 45.3 TFLOPS | Not supported | MI250 only |
| Platform | |||
| Interconnect | Infinity Fabric 400 GB/s | NVLink 112 GB/s | 3.6x higher on MI250 |
| Board power | 500 W | 350 W | 43% higher on MI250 |
| Released | Q4 2021 | Q3 2020 | 13 months newer on MI250 |
| Precision | |||
| Supported formats | FP16BF16FP32FP64INT4INT8 | FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI250 datasheet and RTX 3090 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 | MI250 | RTX 3090 |
|---|---|---|
Cirrascale
USA
|
$1.30 /GPU/hr Reserved | -- |
HyperAI
Hong Kong
|
-- | $0.10 /GPU/hr On-demand |
Vast.ai
USA
|
-- | $0.14 /GPU/hr On-demand |
GPU.ai
UAE
|
-- | $0.15 /GPU/hr On-demand |
Lium
Distributed
|
-- | $0.16 /GPU/hr On-demand |
Runpod
USA
|
-- | $0.22 /GPU/hr On-demand |
Salad
Distributed
|
-- | $0.17 /GPU/hr On-demand |
SwissGPU
Switzerland
|
-- | $0.30 /GPU/hr On-demand |
TensorDock
Distributed
|
-- | $0.31 /GPU/hr On-demand |
1 provider list the MI250 and 9 the RTX 3090. Every listing is on the MI250 page and the RTX 3090 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia A100
80 GB HBM2e · 40% less bandwidth than the MI250
From $0.37 /GPU/hr Compare
-
AMD MI300X
192 GB HBM3 · 66% more bandwidth than the MI250
From $2.59 /GPU/hr Compare
-
Nvidia V100
32 GB HBM2 · 72% less bandwidth than the MI250
From $0.09 /GPU/hr Compare
-
Nvidia RTX 5090
32 GB GDDR7 · 91% more bandwidth than the RTX 3090
From $0.35 /GPU/hr Compare
-
Nvidia RTX 4090
24 GB GDDR6X · PCIe only
From $0.33 /GPU/hr Compare
Our data for AMD MI250 was last updated on Sept. 14, 2026, and for Nvidia RTX 3090 on Sept. 14, 2026.