NVIDIA A100 SXM4 80GB vs NVIDIA RTX A6000 48GB

The A100 delivers 2.0x the FP16 throughput of the RTX A6000 (312 vs 155 TFLOPS dense). The A100 also carries 32 GB more memory (80 GB vs 48 GB).

A100: Ampere, 2020 RTX A6000: Ampere, 2020
FP16 dense lead
2.0x
A100: 312 vs 155 TFLOPS
VRAM
80 vs 48 GB
32 GB more for the A100
Bandwidth
2.0 TB/s vs 768 GB/s
A100 moves data faster
TDP
400 W vs 300 W
RTX A6000 draws 100 W less

Specifications

A100
RTX A6000
Architecture
AmpereAmpere
Launch Year
20202020
Form Factor
SXMPCIe
VRAM
80 GB48 GB
Memory Bandwidth
2.0 TB/s768 GB/s
TDP
400 W300 W
Process Node
7nm8nm

Performance (TFLOPS)

A100
RTX A6000
FP64
9.7 TFlops 0.6 TFlops
FP32
19.5 TFlops 38.7 TFlops
FP16
312.0 TFlops
624.0 TFLOPS sparse
154.8 TFlops
309.6 TFLOPS sparse
BF16
312.0 TFlops
624.0 TFLOPS sparse
154.8 TFlops
309.6 TFLOPS sparse
INT8
624.0 TFlops
1248.0 TOPS sparse
309.7 TFlops
619.4 TOPS sparse

FLOPS by Precision

What actually differs

Both GPUs launched in 2020: the A100 on Ampere and the RTX A6000 on Ampere.

Dense throughput for the A100 against the RTX A6000: FP64 9.7 vs 0.6 TFLOPS, FP32 20 vs 39 TFLOPS, FP16 312 vs 155 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 80 GB against 48 GB, fed at 2.0 TB/s versus 768 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 400 W for the A100 and 300 W for the RTX A6000. At FP32 that works out to 0.05 against 0.13 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 A100 faster than the RTX A6000?

At FP16 precision the A100 reaches 312 TFLOPS dense against 155 TFLOPS for the RTX A6000. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the A100 or the RTX A6000?

The A100 carries 80 GB of VRAM versus 48 GB for the RTX A6000. Memory bandwidth is 2.0 TB/s for the A100 and 768 GB/s for the RTX A6000.

How much power do the A100 and the RTX A6000 draw?

The A100 is rated at 400 W TDP and the RTX A6000 at 300 W. On FP32 throughput per watt, the RTX A6000 is the more efficient part.

Can I rent the A100 or the RTX A6000 in the cloud?

Yes. Live cloud listings tracked by Flopper start at $0.30 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA A100 SXM4 80GB

ProviderInstancePrice/GPU-hrChecked
DataCrunch
1A100.22V (spot) $0.90 3h ago View →
CoreWeave
nvidia-a100 (spot) $1.21 3h ago View →
Massed Compute
gpu_1x_a100_SXM4 $1.38 3h ago View →
RunPod
A100 SXM (Community Cloud) $1.39 3h ago View →
Denvr Dataworks
A100_40GB_SXM_8x $1.40 3h ago View →
Denvr Dataworks
A100_80GB_SXM_1x $1.40 3h ago View →
RunPod
A100 SXM (Secure Cloud) $1.59 3h ago View →
Hyperstack
$1.60 3h ago View →
DataCrunch
1A100.22V (on-demand) $1.79 3h ago View →
Lambda Labs
gpu_1x_a100_sxm4 $1.99 3h ago View →
Crusoe
$2.30 3h ago View →
CoreWeave
nvidia-a100 (on-demand) $2.70 3h ago View →
Lambda Labs
gpu_8x_a100_80gb_sxm4 $2.79 3h ago View →
Excess Supply
gpu_8x_a100_80gb_sxm4 $3.00 3h ago View →
Paperspace
A100 $3.19 3h ago View →
Paperspace
A100-80G $3.28 3h ago View →
Microsoft Azure
ND96ams A100 v4 $4.10 3h ago View →
Microsoft Azure
ND96amsr A100 v4 $4.10 3h ago View →
All A100 listings and price history →

NVIDIA RTX A6000 48GB

ProviderInstancePrice/GPU-hrChecked
DataCrunch
1A6000.10V (spot) $0.30 3h ago View →
RunPod
RTX A6000 (Community Cloud) $0.33 3h ago View →
Hyperstack
$0.40 3h ago View →
Hyperstack
$0.50 3h ago View →
RunPod
RTX A6000 (Secure Cloud) $0.53 3h ago View →
Massed Compute
gpu_1x_A6000_high_perf $0.57 3h ago View →
DataCrunch
1A6000.10V (on-demand) $0.61 3h ago View →
Paperspace
A6000 $1.93 3h ago View →
All RTX A6000 listings and price history →

Get Comparison Updates

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