Two Nvidia data center GPUs, one Ampere and one Pascal: 48 GB GDDR6 against 8 GB GDDR5 and 3.6x the memory bandwidth on the A40.

At a glance

Updated 8 minutes ago
Memory
48 GB
GDDR6
Memory bandwidth
696 GB/s
From
$1.05 /GPU/hr
6 providers
Memory
8 GB
GDDR5
Memory bandwidth
192 GB/s
From
$0.65 /GPU/hr
1 provider

Price comparison

Cheapest listed price per GPU per hour by billing type, and the median across providers.

Billing A40 P4 Difference
On-demand $1.05 $0.65 37% cheaper on P4
Spot No spot listings $0.12
Reserved $0.61 $0.34 44% cheaper on P4
Median $1.05 $0.81 22% cheaper on P4

Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.

Price history

A40 vs P4 price history

Weekly median on-demand price per GPU per hour

Last 12 months
Aggregating historical prices...

Specs comparison

Spec A40 P4 Difference
Memory
Capacity 48 GB GDDR6 8 GB GDDR5 6x more on A40
Bandwidth 696 GB/s 192 GB/s 3.6x higher on A40
Bus width 384-bit 256-bit 1.5x wider on A40
Compute · dense
FP16 / BF16 149.7 TFLOPS
INT8 299.3 TOPS 22 TOPS 14x higher on A40
FP32 37.4 TFLOPS 5.5 TFLOPS 6.8x higher on A40
Platform
Interconnect PCIe only PCIe only
Board power 300 W 75 W 4x higher on A40
Released Q4 2020 Q3 2016 4 years newer on A40
Precision
Supported formats FP16BF16TF32FP32INT4INT8 FP16FP32INT8

Compute: Vendor peak figures, dense and per GPU. Not measured throughput.

Source: A40 datasheet and P4 datasheet.

Where to rent them

Providers listing these GPUs, those with both first, so you can switch without moving clouds.

Provider A40 /GPU/hr P4 /GPU/hr
Exoscale logo Exoscale On-Demand $1.05 --
Database Mart logo Database Mart Reserved (24mo) $0.61 --
Utho logo Utho Reserved (36mo) $0.65 --
Runpod logo Runpod Reservation on request --
Sesterce logo Sesterce On-Demand $1.21 --
EmpirioLabs AI logo EmpirioLabs AI On-Demand $1.49 --
Google Cloud logo Google Cloud -- On-Demand $0.65

Similar GPUs

Other comparisons in the same class.

Our data for Nvidia A40 was last updated on Sept. 7, 2026, and for Nvidia P4 on Sept. 7, 2026.