A data center GPU against a consumer card, one Kepler and one Blackwell: 12 GB GDDR5 against 32 GB GDDR7 and 7.5x the memory bandwidth on the RTX 5090.
At a glance
Updated 6 minutes ago- Memory
- 12 GB
- GDDR5
- Memory bandwidth
- 240 GB/s
- From
- $0.90 /GPU/hr
- 2 providers
- Memory
- 32 GB
- GDDR7
- Memory bandwidth
- 1.79 TB/s
- From
- $0.35 /GPU/hr
- 18 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | K80 | RTX 5090 | Difference |
|---|---|---|---|
| On-demand | $0.90 | $0.35 | 2.6x more expensive on K80 |
| Spot | No spot listings | $0.25 | |
| Reserved | $0.04 | $0.21 | 4.7x more expensive on RTX 5090 |
| Median | $0.90 | $0.63 | 30% cheaper on RTX 5090 |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
K80 vs RTX 5090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | K80 | RTX 5090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 12 GB GDDR5 | 32 GB GDDR7 | 2.7x more on RTX 5090 |
| Bandwidth | 240 GB/s | 1,792 GB/s | 7.5x higher on RTX 5090 |
| Bus width | 384-bit | 512-bit | 1.3x wider on RTX 5090 |
| Compute · dense | |||
| FP4 | Not supported | 1,676 TFLOPS | RTX 5090 only |
| FP8 | Not supported | 419 TFLOPS | RTX 5090 only |
| FP16 / BF16 | Not supported | 209.5 TFLOPS | RTX 5090 only |
| INT8 | Not supported | 838 TOPS | RTX 5090 only |
| FP32 | 4.4 TFLOPS | 104.8 TFLOPS | 24x higher on RTX 5090 |
| FP64 | 1.5 TFLOPS | Not supported | K80 only |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 150 W | 575 W | 3.8x higher on RTX 5090 |
| Released | Q4 2014 | Q1 2025 | 10 years newer on RTX 5090 |
| Precision | |||
| Supported formats | FP32FP64 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: K80 datasheet and RTX 5090 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | K80 /GPU/hr | RTX 5090 /GPU/hr |
|---|---|---|
AWS
|
On-Demand $0.90 | -- |
Database Mart
|
Reserved (24mo) $0.0447 | -- |
HyperAI
|
-- | On-Demand $0.35 |
Salad
|
-- | On-Demand $0.45 |
GPUhub
|
-- | On-Demand $0.46 |
Nova Cloud
|
-- | On-Demand $0.63 |
Vast.ai
|
-- | On-Demand $0.73 |
GPU.ai
|
-- | On-Demand $53.34 |
GPUaaS
|
-- | Custom on request |
SwissGPU
|
-- | On-Demand $0.61 |
2 providers list the K80 and 18 the RTX 5090. Every listing is on the K80 page and the RTX 5090 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia K80 was last updated on Sept. 7, 2026, and for Nvidia RTX 5090 on Sept. 7, 2026.