A data center GPU against a consumer card, one Grace Blackwell and one Ada Lovelace: 186 GB HBM3e against 24 GB GDDR6X, 7.9x the memory bandwidth on the GB200, and NVLink 1,800 GB/s on the GB200 where the RTX 4090 is PCIe only.
At a glance
Memory
- GB200
- 186 GB
- RTX 4090
- 24 GB
7.8x more on GB200
Memory bandwidth
- GB200
- 8.00 TB/s
- RTX 4090
- 1.01 TB/s
7.9x higher on GB200
Cheapest on-demand
- GB200
- $10.50 /GPU/hr
- RTX 4090
- $0.23 /GPU/hr
45x more expensive on GB200
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | GB200 | RTX 4090 | Difference |
|---|---|---|---|
| On-demand | $16.00 /GPU/hr from $10.50 | $0.44 /GPU/hr from $0.23 | 37x more expensive on GB200 |
| Reserved | -- /GPU/hr from $10.58 | $0.46 /GPU/hr from $0.27 (12 mo) | |
| Spot | -- /GPU/hr from $27.04 | -- /GPU/hr from $0.16 |
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 GB200 and spot on the RTX 4090: 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
GB200 vs RTX 4090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | GB200 | RTX 4090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 186 GB HBM3e | 24 GB GDDR6X | 7.8x more on GB200 |
| Bandwidth | 8,000 GB/s | 1,010 GB/s | 7.9x higher on GB200 |
| Bus width | 384-bit | ||
| Compute · dense | |||
| FP4 | 10,000 TFLOPS | Not supported | GB200 only |
| FP8 | 5,000 TFLOPS | 330.3 TFLOPS | 15x higher on GB200 |
| FP16 / BF16 | 2,500 TFLOPS | 165.2 TFLOPS | 15x higher on GB200 |
| INT8 | 5,000 TOPS | 660.6 TOPS | 7.6x higher on GB200 |
| FP32 | 80 TFLOPS | 82.6 TFLOPS | Within 5% |
| FP64 | 40 TFLOPS | Not supported | GB200 only |
| Platform | |||
| Interconnect | NVLink 1,800 GB/s | PCIe only | NVLink on GB200 only |
| Board power | 1,200 W | 450 W | 2.7x higher on GB200 |
| Released | Q1 2024 | Q4 2022 | 17 months newer on GB200 |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: GB200 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 | GB200 /GPU/hr | RTX 4090 /GPU/hr |
|---|---|---|
CoreWeave
|
On-Demand $10.50 | -- |
Oracle Cloud
|
On-Demand $16.00 | -- |
Azure
|
On-Demand $27.04 | -- |
AWS
|
Reservation $10.58 | -- |
CUDO
|
Custom on request | -- |
Canopy Wave
|
Custom on request | -- |
Nebius
|
Custom on request | -- |
Together AI
|
Custom on request | -- |
Theta EdgeCloud
|
-- | On-Demand $0.23 |
Salad
|
-- | On-Demand $0.30 |
Novita
|
-- | On-Demand $0.33 |
GPU.ai
|
-- | On-Demand $0.34 |
Runpod
|
-- | On-Demand $0.34 |
Packet·ai
|
-- | On-Demand $0.39 |
Vast.ai
|
-- | On-Demand $0.40 |
Lium
|
-- | On-Demand $0.58 |
8 providers list the GB200 and 18 the RTX 4090. Every listing is on the GB200 page and the RTX 4090 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia GB200 was last updated on Sept. 7, 2026, and for Nvidia RTX 4090 on Sept. 7, 2026.