AMD MI325X
vs
Nvidia V100
The MI325X is newer, with 8x the memory and 6.7x the bandwidth. The V100 is typically cheaper.
At a glance
Memory
8x more on MI325X
- MI325X
- 256 GB
- V100
- 32 GB
Memory bandwidth
6.7x higher on MI325X
- MI325X
- 6,000 GB/s
- V100
- 900 GB/s
Cheapest on-demand
94% cheaper on V100
- MI325X
- $2.25 /GPU/hr
- V100
- $0.14 /GPU/hr
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | MI325X | V100 | Difference |
|---|---|---|---|
| On-demand | $3.11 /GPU/hr from $2.25 | $1.38 /GPU/hr from $0.14 | 55% cheaper on V100 |
| Reserved | $1.95 /GPU/hr from $1.50 (12 mo) | $1.51 /GPU/hr from $0.17 (24 mo) | 23% cheaper on V100 |
| Spot | Sold out | $0.40 /GPU/hr from $0.11 |
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
MI325X vs V100 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | MI325X | V100 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 256 GB HBM3e | 32 GB HBM2 | 8x more on MI325X |
| Bandwidth | 6,000 GB/s | 900 GB/s | 6.7x higher on MI325X |
| Bus width | 8,192-bit | 4,096-bit | 2x wider on MI325X |
| Compute · dense | |||
| FP8 | 2,614.9 TFLOPS | Not supported | MI325X only |
| FP16 / BF16 | 1,307.4 TFLOPS | 125 TFLOPS | 10x higher on MI325X |
| INT8 | 2,614.9 TOPS | ||
| FP32 | 163.4 TFLOPS | 15.7 TFLOPS | 10x higher on MI325X |
| FP64 | 81.7 TFLOPS | 7.8 TFLOPS | 10x higher on MI325X |
| Platform | |||
| Interconnect | Infinity Fabric 896 GB/s | NVLink 300 GB/s | 3x higher on MI325X |
| Board power | 1,000 W | 300 W | 3.3x higher on MI325X |
| Released | Q4 2024 | Q2 2017 | 7 years newer on MI325X |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32FP64INT8 | FP16FP32FP64INT8 | |
Figures are for each model's headline configuration. Where a form factor carries a different number, it is named under the figure.
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI325X datasheet and V100 datasheet.
Where to rent them
Providers that rent these GPUs, with each one's cheapest rate. Providers that rent both are listed first.
| Provider | MI325X | V100 |
|---|---|---|
Cyfuture AI
India
|
$3.06 /GPU/hr On-demand | $0.55 /GPU/hr On-demand |
Bentaus
USA
|
$2.25 /GPU/hr On-demand | -- |
DigitalOcean
USA
|
$3.80 /GPU/hr On-demand | -- |
TensorWave
USA
|
$2.25 /GPU/hr Custom OAM | -- |
Vultr
USA
|
$2.00 /GPU/hr Spot | -- |
HyperAI
Hong Kong
|
-- | $0.14 /GPU/hr On-demand |
Verda
Finland
|
-- | $0.23 /GPU/hr On-demand |
Sesterce
France
|
-- | $0.43 /GPU/hr On-demand |
Runcrate
USA
|
-- | $0.51 /GPU/hr On-demand NVL |
Geodd
USA
|
-- | $0.16 /GPU/hr On-demand |
TensorDock
Distributed
|
-- | $0.28 /GPU/hr On-demand SXM |
OVH
France
|
-- | $0.77 /GPU/hr On-demand |
5 providers list the MI325X and 19 the V100. Every listing is on the MI325X page and the V100 page.
Similar GPUs
Other comparisons in the same class.
Our data for AMD MI325X was last updated on Sept. 14, 2026, and for Nvidia V100 on Sept. 14, 2026.