Skip to main content

A10G vs L4

The A10G delivers 1.0× the FP16 compute of the L4 while costing 1.7× more on average, making the L4 65% more cost-efficient per TFLOP.

Spec Comparison

Spec
A10G
L4
Architecture
Ampere
Ada Lovelace
VRAM
24 GB
24 GB
FP16 TFLOPS
125
122
Memory Bandwidth
600 GB/s
300 GB/s
MSRP
$3,500
$2,500
Used/Market Price
$2,000
$1,800

Live Pricing

A10G
Cheapest
$0.342/hr
Median
$2.13/hr
Average
$4.35/hr
Providers
5
Instances
134
L4
Cheapest
$0.188/hr
Median
$1.26/hr
Average
$2.50/hr
Providers
9
Instances
374

Cost Efficiency

Metric
A10G
L4
$/TFLOP/hr
$0.0171
$0.0104
$/GB-VRAM/hr
$0.0889
$0.0526

Lower $/TFLOP and $/GB-VRAM is better. Based on median hourly pricing.

Provider Availability

Both available on (3)

Only on A10G (2)

A10G Cloud Pricing

134 instances
ProviderGPUsVRAMRegionTypePrice/hr
CHEAPESTAWS1x22.4 GBUS-EastSpot$0.342Rent
AWS1x22.4 GBAsia-PacificSpot$0.391Rent
AWS1x22.4 GBAsia-PacificSpot$0.415Rent
AWS1x22.4 GBSouth AmericaSpot$0.416Rent
AWS1x22.4 GBeu-north-1Spot$0.420Rent
AWS1x22.4 GBeu-north-1Spot$0.422Rent
AWS1x22.4 GBeu-north-1Spot$0.468Rent
AWS1x22.4 GBSouth AmericaSpot$0.469Rent
AWS1x22.4 GBUS-WestSpot$0.496Rent
AWS1x22.4 GBAsia-PacificSpot$0.504Rent

L4 Cloud Pricing

374 instances
ProviderGPUsVRAMRegionTypePrice/hr
CHEAPESTAWS1x24 GBUS-EastSpot$0.188Rent
GCP1x24 GBAsia-PacificSpot$0.192Rent
RunPod1x24 GBEUR-IS-1Spot$0.220Rent
RunPod1x24 GBUS-MO-2Spot$0.220Rent
GCP1x24 GBAsia-PacificSpot$0.229Rent
AWS1x24 GBUS-EastSpot$0.250Rent
AWS1x24 GBEU-WestSpot$0.261Rent
GCP1x24 GBAsia-PacificSpot$0.265Rent
AWS1x24 GBAsia-PacificSpot$0.272Rent
AWS1x24 GBAsia-PacificSpot$0.274Rent

Compare with Other GPUs