Skip to content
GPUZone

INFERENTIA 2 vs L40S

The L40S has 48 GB of memory against 32 GB on the INFERENTIA 2.

MeasureAWS Inferentia2NVIDIA L40S 48GB
Lowest per GPU-hour—$0.38
Median per GPU-hour—$1.30
On-demand from—$0.38
Spot from—$0.45
Reserved from—$0.49
Lowest per A100-equivalent hour—$0.40
Live offers0284
Providers085
Memory32 GB HBM2e48 GB GDDR6
Memory bandwidth820 GB/s864 GB/s
FP16 / BF16190 TFLOPS362 TFLOPS
FP8190 TFLOPS733 TFLOPS
Performance vs A100 80GB0.52×0.96×
ArchitectureInferentia2 (NeuronCore v2)Ada Lovelace
Power—350 W

Where the offers sit

AWS Inferentia2

No live offers right now.

NVIDIA L40S 48GB