Two Nvidia data center GPUs, one Ampere and one Blackwell: 48 GB GDDR6 against up to 192 GB HBM3e, 11x the memory bandwidth on the B100, and NVLink 1,800 GB/s on the B100 where the A40 is PCIe only.
At a glance
Updated 8 minutes ago- Memory
- 48 GB
- GDDR6
- Memory bandwidth
- 696 GB/s
- From
- $1.05 /GPU/hr
- 6 providers
- Memory
- 192 GB
- HBM3e
- Memory bandwidth
- 8.00 TB/s
- From
- --
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | A40 | B100 | Difference |
|---|---|---|---|
| On-demand | $1.05 | -- | |
| Reserved | $0.61 | No reserved listings | |
| Median | $1.05 | -- |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
A40 vs B100 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | A40 | B100 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 48 GB GDDR6 | 192 GB HBM3e | 4x more on B100 |
| Bandwidth | 696 GB/s | 8,000 GB/s | 11x higher on B100 |
| Bus width | 384-bit | ||
| Compute · dense | |||
| FP4 | Not supported | 7,000 TFLOPS | B100 only |
| FP8 | Not supported | 3,500 TFLOPS | B100 only |
| FP16 / BF16 | 149.7 TFLOPS | 1,750 TFLOPS | 12x higher on B100 |
| INT8 | 299.3 TOPS | 3,500 TOPS | 12x higher on B100 |
| FP32 | 37.4 TFLOPS | 60 TFLOPS | 1.6x higher on B100 |
| FP64 | Not supported | 30 TFLOPS | B100 only |
| Platform | |||
| Interconnect | PCIe only | NVLink 1,800 GB/s | NVLink on B100 only |
| Released | Q4 2020 | Q4 2024 | 4 years newer on B100 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: A40 datasheet and B100 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | A40 /GPU/hr | B100 /GPU/hr |
|---|---|---|
Exoscale
|
On-Demand $1.05 | -- |
Database Mart
|
Reserved (24mo) $0.61 | -- |
Utho
|
Reserved (36mo) $0.65 | -- |
Runpod
|
Reservation on request | -- |
Sesterce
|
On-Demand $1.21 | -- |
EmpirioLabs AI
|
On-Demand $1.49 | -- |
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia A40 was last updated on Sept. 7, 2026, and for Nvidia B100 on Sept. 7, 2026.