AMD MI300X
vs
Nvidia RTX 4090
A data center GPU against a consumer card, one CDNA 3 and one Ada Lovelace: 192 GB HBM3 against 24 GB GDDR6X, 5.2x the memory bandwidth on the MI300X, and Infinity Fabric 896 GB/s on the MI300X where the RTX 4090 is PCIe only.
At a glance
Updated 7 minutes ago- Memory
- 192 GB
- HBM3
- Memory bandwidth
- 5.30 TB/s
- From
- $2.39 /GPU/hr
- 10 providers
- Memory
- 24 GB
- GDDR6X
- Memory bandwidth
- 1.01 TB/s
- From
- $0.23 /GPU/hr
- 18 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | MI300X | RTX 4090 | Difference |
|---|---|---|---|
| On-demand | $2.39 | $0.23 | 10x more expensive on MI300X |
| Spot | $1.60 | $0.16 | 10x more expensive on MI300X |
| Reserved | $1.61 | $0.27 | 5.9x more expensive on MI300X |
| Median | $2.91 | $0.44 | 6.6x more expensive on MI300X |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
MI300X vs RTX 4090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | MI300X | RTX 4090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 192 GB HBM3 | 24 GB GDDR6X | 8x more on MI300X |
| Bandwidth | 5,300 GB/s | 1,010 GB/s | 5.2x higher on MI300X |
| Bus width | 8,192-bit | 384-bit | 21x wider on MI300X |
| Compute · dense | |||
| FP8 | 2,614.9 TFLOPS | 330.3 TFLOPS | 7.9x higher on MI300X |
| FP16 / BF16 | 1,307.4 TFLOPS | 165.2 TFLOPS | 7.9x higher on MI300X |
| INT8 | 2,614.9 TOPS | 660.6 TOPS | 4x higher on MI300X |
| FP32 | 163.4 TFLOPS | 82.6 TFLOPS | 2x higher on MI300X |
| FP64 | 81.7 TFLOPS | Not supported | MI300X only |
| Platform | |||
| Interconnect | Infinity Fabric 896 GB/s | PCIe only | Infinity Fabric on MI300X only |
| Board power | 750 W | 450 W | 1.7x higher on MI300X |
| Released | Q4 2023 | Q4 2022 | 14 months newer on MI300X |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32FP64INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: MI300X datasheet and RTX 4090 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | MI300X /GPU/hr | RTX 4090 /GPU/hr |
|---|---|---|
Runpod
|
On-Demand $2.39 | On-Demand $0.34 |
DigitalOcean
|
On-Demand $2.59 | -- |
Cyfuture AI
|
On-Demand $2.78 | -- |
Hot Aisle
|
On-Demand $2.99 | -- |
Oracle Cloud
|
On-Demand $6.00 | -- |
Azure
|
On-Demand $7.86 | -- |
Cirrascale
|
Reserved (12mo) $3.12 | -- |
Crusoe
|
Reservation on request | -- |
Theta EdgeCloud
|
-- | On-Demand $0.23 |
Salad
|
-- | On-Demand $0.30 |
Novita
|
-- | On-Demand $0.33 |
GPU.ai
|
-- | On-Demand $0.34 |
Packet·ai
|
-- | On-Demand $0.39 |
Lium
|
-- | On-Demand $0.58 |
Sesterce
|
-- | On-Demand $0.66 |
10 providers list the MI300X and 18 the RTX 4090. Every listing is on the MI300X page and the RTX 4090 page.
Similar GPUs
Other comparisons in the same class.
Our data for AMD MI300X was last updated on Sept. 7, 2026, and for Nvidia RTX 4090 on Sept. 7, 2026.