AMD MI350X
vs
Nvidia GH200
Two data center GPUs, one CDNA 4 and one Grace Hopper: 288 GB HBM3e against 96 GB HBM3 and 2x the memory bandwidth on the MI350X.
At a glance
Memory
- MI350X
- 288 GB
- GH200
- 96 GB
3x more on MI350X
Memory bandwidth
- MI350X
- 8.00 TB/s
- GH200
- 4.00 TB/s
2x higher on MI350X
Cheapest on-demand
- MI350X
- --
- GH200
- $6.50 /GPU/hr
MI350X is not sold on demand
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | MI350X | GH200 | Difference |
|---|---|---|---|
| On-demand | No on-demand listings | -- /GPU/hr from $6.50 | |
| Reserved | -- /GPU/hr from $4.76 (12 mo) | No reserved listings | |
| Spot | -- /GPU/hr from $4.00 | Sold out |
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 and spot on the MI350X and on-demand on the GH200: 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
MI350X vs GH200 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | MI350X | GH200 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 288 GB HBM3e | 96 GB HBM3 | 3x more on MI350X |
| Bandwidth | 8,000 GB/s | 4,000 GB/s | 2x higher on MI350X |
| Bus width | 8,192-bit | ||
| Compute · dense | |||
| FP4 | 9,227.5 TFLOPS | Not supported | MI350X only |
| FP8 | 4,614 TFLOPS | 1,979 TFLOPS | 2.3x higher on MI350X |
| FP16 / BF16 | 2,309.6 TFLOPS | 989.5 TFLOPS | 2.3x higher on MI350X |
| INT8 | 4,614 TOPS | 1,979 TOPS | 2.3x higher on MI350X |
| FP32 | 144.2 TFLOPS | 67 TFLOPS | 2.2x higher on MI350X |
| FP64 | 72.1 TFLOPS | 34 TFLOPS | 2.1x higher on MI350X |
| Platform | |||
| Interconnect | Infinity Fabric 1,075 GB/s | NVLink 900 GB/s | 1.2x higher on MI350X |
| Released | Q2 2025 | Q2 2023 | 2 years newer on MI350X |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16FP32FP64INT8 | FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI350X datasheet and GH200 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | MI350X /GPU/hr | GH200 /GPU/hr |
|---|---|---|
DigitalOcean
|
Reserved (12mo) $4.76 | -- |
CoreWeave
|
-- | On-Demand $6.50 |
Lambda Labs
|
-- | On-Demand $2.29 |
Sesterce
|
-- | On-Demand $2.52 |
Vultr
|
-- | Spot $1.99 |
Similar GPUs
Other comparisons in the same class.
-
Nvidia B200
180 GB HBM3e · released Q1 2024
From $3.75 /GPU/hr Compare
-
Nvidia H200
141 GB HBM3e · 40% less bandwidth than the MI350X
From $2.09 /GPU/hr Compare
-
AMD MI355X
288 GB HBM3e · released Q2 2025
From $8.60 /GPU/hr Compare
-
Nvidia H100
80 GB HBM3 · 16% less bandwidth than the GH200
From $1.30 /GPU/hr Compare
Our data for AMD MI350X was last updated on Sept. 14, 2026, and for Nvidia GH200 on Sept. 14, 2026.