A consumer GPU against a workstation card, one Ampere and one Ada Lovelace: 24 GB GDDR6X against 48 GB GDDR6, memory bandwidth within 5%, and NVLink 112 GB/s on the RTX 3090 where the RTX 6000 Ada is PCIe only.
At a glance
Updated 5 minutes ago- Memory
- 24 GB
- GDDR6X
- Memory bandwidth
- 936 GB/s
- From
- $0.10 /GPU/hr
- 8 providers
- Memory
- 48 GB
- GDDR6
- Memory bandwidth
- 960 GB/s
- From
- $0.86 /GPU/hr
- 10 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | RTX 3090 | RTX 6000 Ada | Difference |
|---|---|---|---|
| On-demand | $0.10 | $0.86 | 8.6x more expensive on RTX 6000 Ada |
| Spot | $0.09 | No spot listings | |
| Reserved | $0.06 | $1.04 | 17x more expensive on RTX 6000 Ada |
| Median | $0.31 | $1.23 | 4x more expensive on RTX 6000 Ada |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
RTX 3090 vs RTX 6000 Ada price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | RTX 3090 | RTX 6000 Ada | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 24 GB GDDR6X | 48 GB GDDR6 | 2x more on RTX 6000 Ada |
| Bandwidth | 936 GB/s | 960 GB/s | Within 5% |
| Bus width | 384-bit | 384-bit | Identical |
| Compute · dense | |||
| FP8 | Not supported | 728.5 TFLOPS | RTX 6000 Ada only |
| FP16 / BF16 | 71.2 TFLOPS | 364.2 TFLOPS | 5.1x higher on RTX 6000 Ada |
| INT8 | 284.7 TOPS | ||
| FP32 | 35.6 TFLOPS | 91.1 TFLOPS | 2.6x higher on RTX 6000 Ada |
| Platform | |||
| Interconnect | NVLink 112 GB/s | PCIe only | NVLink on RTX 3090 only |
| Board power | 350 W | 300 W | 1.2x higher on RTX 3090 |
| Released | Q3 2020 | Q4 2022 | 2 years newer on RTX 6000 Ada |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: RTX 3090 datasheet and RTX 6000 Ada datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | RTX 3090 /GPU/hr | RTX 6000 Ada /GPU/hr |
|---|---|---|
TensorDock
|
On-Demand $0.31 | On-Demand $0.86 |
SwissGPU
|
On-Demand $0.31 | On-Demand $1.23 |
Runpod
|
Reservation on request | Reservation on request |
HyperAI
|
On-Demand $0.10 | -- |
Theta EdgeCloud
|
On-Demand $0.13 | -- |
Salad
|
On-Demand $0.25 | -- |
GPU.ai
|
On-Demand $0.22 | -- |
EmpirioLabs AI
|
On-Demand $0.40 | -- |
DigitalOcean
|
-- | On-Demand $1.57 |
AceCloud
|
-- | Reserved (12mo) $1.04 |
Sesterce
|
-- | On-Demand $0.66 |
Novita
|
-- | On-Demand $0.77 |
Massed Compute
|
-- | On-Demand $0.79 |
8 providers list the RTX 3090 and 10 the RTX 6000 Ada. Every listing is on the RTX 3090 page and the RTX 6000 Ada page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia RTX 3090 was last updated on Sept. 7, 2026, and for Nvidia RTX 6000 Ada on Sept. 7, 2026.