Two Nvidia consumer GPUs on the Ampere architecture: 12 GB GDDR6X against 24 GB GDDR6X, memory bandwidth within 5%, and NVLink 112 GB/s on the RTX 3090 where the RTX 3080 Ti is PCIe only.
At a glance
Updated 10 minutes ago- Memory
- 12 GB
- GDDR6X
- Memory bandwidth
- 912 GB/s
- From
- $0.20 /GPU/hr
- 3 providers
- Memory
- 24 GB
- GDDR6X
- Memory bandwidth
- 936 GB/s
- From
- $0.10 /GPU/hr
- 7 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | RTX 3080 Ti | RTX 3090 | Difference |
|---|---|---|---|
| On-demand | $0.20 | $0.10 | 2x more expensive on RTX 3080 Ti |
| Spot | $0.08 | $0.09 | 11% cheaper on RTX 3080 Ti |
| Reserved | No reserved listings | $0.06 | |
| Median | $0.53 | $0.31 | 41% cheaper on RTX 3090 |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
RTX 3080 Ti vs RTX 3090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | RTX 3080 Ti | RTX 3090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 12 GB GDDR6X | 24 GB GDDR6X | 2x more on RTX 3090 |
| Bandwidth | 912 GB/s | 936 GB/s | Within 5% |
| Bus width | 384-bit | 384-bit | Identical |
| Compute · dense | |||
| FP16 / BF16 | 68.2 TFLOPS | 71.2 TFLOPS | Within 5% |
| INT8 | 272.8 TOPS | 284.7 TOPS | Within 5% |
| FP32 | 34.1 TFLOPS | 35.6 TFLOPS | Within 5% |
| Platform | |||
| Interconnect | PCIe only | NVLink 112 GB/s | NVLink on RTX 3090 only |
| Board power | 350 W | 350 W | Identical |
| Released | Q2 2021 | Q3 2020 | 8 months newer on RTX 3080 Ti |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: RTX 3080 Ti datasheet and RTX 3090 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | RTX 3080 Ti /GPU/hr | RTX 3090 /GPU/hr |
|---|---|---|
Salad
|
On-Demand $0.20 | On-Demand $0.25 |
Runpod
|
On-Demand $0.18 | Reservation on request |
Exoscale
|
On-Demand $0.79 | -- |
HyperAI
|
-- | On-Demand $0.10 |
Theta EdgeCloud
|
-- | On-Demand $0.13 |
SwissGPU
|
-- | On-Demand $0.31 |
TensorDock
|
-- | On-Demand $0.31 |
EmpirioLabs AI
|
-- | On-Demand $0.40 |
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia RTX 3080 Ti was last updated on Sept. 7, 2026, and for Nvidia RTX 3090 on Sept. 7, 2026.