AMD MI325X
vs
Nvidia B200
The B200 has 28% more bandwidth, while the MI325X has 42% more memory. The MI325X is typically cheaper.
At a glance
Memory
42% more on MI325X
- MI325X
- 256 GB
- B200
- 180 GB
Memory bandwidth
28% higher on B200
- MI325X
- 6.00 TB/s
- B200
- 7.70 TB/s
Cheapest on-demand
40% cheaper on MI325X
- MI325X
- $2.25 /GPU/hr
- B200
- $3.75 /GPU/hr
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | MI325X | B200 | Difference |
|---|---|---|---|
| On-demand | $3.09 /GPU/hr from $2.25 | $8.19 /GPU/hr from $3.75 | 62% cheaper on MI325X |
| Reserved | $1.95 /GPU/hr from $1.49 (12 mo) | $6.71 /GPU/hr from $3.75 | 71% cheaper on MI325X |
| Spot | Sold out | $4.26 /GPU/hr from $3.59 |
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 B200 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | MI325X | B200 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 256 GB HBM3e | 180 GB HBM3e | 42% more on MI325X |
| Bandwidth | 6,000 GB/s | 7,700 GB/s | 28% higher on B200 |
| Bus width | 8,192-bit | ||
| Compute · dense | |||
| FP4 | Not supported | 9,000 TFLOPS | B200 only |
| FP8 | 2,614.9 TFLOPS | 4,500 TFLOPS | 72% higher on B200 |
| FP16 / BF16 | 1,307.4 TFLOPS | 2,250 TFLOPS | 72% higher on B200 |
| INT8 | 2,614.9 TOPS | 4,500 TOPS | 72% higher on B200 |
| FP32 | 163.4 TFLOPS | 75 TFLOPS | 2.2x higher on MI325X |
| FP64 | 81.7 TFLOPS | 37 TFLOPS | 2.2x higher on MI325X |
| Platform | |||
| Interconnect | Infinity Fabric 896 GB/s | NVLink 1,800 GB/s | 2x higher on B200 |
| Board power | 1,000 W | 1,000 W | Identical |
| Released | Q4 2024 | Q1 2024 | 7 months newer on MI325X |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32FP64INT8 | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI325X datasheet and B200 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 | MI325X | B200 |
|---|---|---|
Bentaus
USA
|
$2.25 /GPU/hr On-demand | -- |
Cyfuture AI
India
|
$3.04 /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 | -- |
Packet·ai
USA
|
-- | $3.75 /GPU/hr On-demand |
Lium
Distributed
|
-- | $5.00 /GPU/hr On-demand SXM |
Daytona
USA
|
-- | $6.25 /GPU/hr On-demand |
Sesterce
France
|
-- | $7.17 /GPU/hr On-demand |
Runpod
USA
|
-- | $7.99 /GPU/hr On-demand SXM |
Verda
Finland
|
-- | $8.44 /GPU/hr On-demand SXM |
Runcrate
USA
|
-- | $8.48 /GPU/hr On-demand NVL |
Lyceum
Germany
|
-- | $8.79 /GPU/hr On-demand |
5 providers list the MI325X and 38 the B200. Every listing is on the MI325X page and the B200 page.
Similar GPUs
Other comparisons in the same class.
Our data for AMD MI325X was last updated on Sept. 14, 2026, and for Nvidia B200 on Sept. 14, 2026.