NVIDIA A10 24GB vs NVIDIA A40 48GB

The A40 delivers 20% more FP16 throughput than the A10 (150 vs 125 TFLOPS dense). The A40 also carries 24 GB more memory (48 GB vs 24 GB).

A10: Ampere, 2021 A40: Ampere, 2021
FP16 dense lead
+20%
A40: 150 vs 125 TFLOPS
VRAM
24 vs 48 GB
24 GB more for the A40
Bandwidth
600 GB/s vs 696 GB/s
A40 moves data faster
TDP
150 W vs 300 W
A10 draws 150 W less

Specifications

A10
A40
Architecture
AmpereAmpere
Launch Year
20212021
Form Factor
PCIePCIe
VRAM
24 GB48 GB
Memory Bandwidth
600 GB/s696 GB/s
TDP
150 W300 W
Process Node

Performance (TFLOPS)

A10
A40
FP32
31.2 TFlops 37.4 TFlops
FP16
125.0 TFlops
250.0 TFLOPS sparse
149.7 TFlops
299.4 TFLOPS sparse
BF16
125.0 TFlops
250.0 TFLOPS sparse
149.7 TFlops
299.4 TFLOPS sparse
INT8
250.0 TFlops
500.0 TOPS sparse
299.3 TFlops
598.6 TOPS sparse

FLOPS by Precision

What actually differs

Both GPUs launched in 2021: the A10 on Ampere and the A40 on Ampere.

Dense throughput for the A10 against the A40: FP32 31 vs 37 TFLOPS, FP16 125 vs 150 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 24 GB against 48 GB, fed at 600 GB/s versus 696 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.

Power budgets are 150 W for the A10 and 300 W for the A40. At FP32 that works out to 0.21 against 0.12 TFLOPS per watt.

For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.

Frequently asked questions

Is the A10 faster than the A40?

At FP16 precision the A40 reaches 150 TFLOPS dense against 125 TFLOPS for the A10. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the A10 or the A40?

The A40 carries 48 GB of VRAM versus 24 GB for the A10. Memory bandwidth is 600 GB/s for the A10 and 696 GB/s for the A40.

How much power do the A10 and the A40 draw?

The A10 is rated at 150 W TDP and the A40 at 300 W. On FP32 throughput per watt, the A10 is the more efficient part.

Can I rent the A10 or the A40 in the cloud?

Yes. Live cloud listings tracked by Flopper start at $0.35 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA A10 24GB

ProviderConfigurationPrice/GPU-hrChecked
Lambda Labs
1× A10 24GB $1.29 4h ago View →
Microsoft Azure
1× A10 24GB $3.20 4h ago View →
Microsoft Azure
2× A10 24GB $3.26 4h ago View →
Microsoft Azure
1× A10 24GB $4.52 4h ago View →
All A10 listings and price history →

NVIDIA A40 48GB

ProviderConfigurationPrice/GPU-hrChecked
RunPod
1× A40 48GB Community $0.35 4h ago View →
RunPod
1× A40 48GB $0.44 4h ago View →
All A40 listings and price history →

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.