NVIDIA A16 16GB vs NVIDIA GeForce RTX 3090 24GB

The GeForce RTX 3090 delivers 4.0x the FP16 throughput of the A16 (71 vs 18 TFLOPS dense). The GeForce RTX 3090 also carries 8 GB more memory (24 GB vs 16 GB).

A16: Ampere, 2021 GeForce RTX 3090: Ampere, 2020
FP16 dense lead
4.0x
GeForce RTX 3090: 71 vs 18 TFLOPS
Memory
16 vs 24 GB
8 GB more for the GeForce RTX 3090
Bandwidth
200 GB/s vs 936 GB/s
GeForce RTX 3090 moves data faster

Specifications

A16
GeForce RTX 3090
Architecture
AmpereAmpere
Launch Year
20212020
Form Factor
PCIePCIe
Memory
16 GB24 GB
Memory Bandwidth
200 GB/s936 GB/s
TDP
350 W
Process Node
8nm

Performance (TFLOPS)

A16
GeForce RTX 3090
FP64
No verified data available0.6 TFlops
FP32
4.5 TFlops 35.6 TFlops
TF32
9.0 TFlops
18.0 TFLOPS sparse
35.6 TFlops
71.0 TFLOPS sparse
BF16
No verified data available71.0 TFlops
142.0 TFLOPS sparse
FP16
17.9 TFlops
35.9 TFLOPS sparse
71.0 TFlops
142.0 TFLOPS sparse
FP8
No verified data available No verified data available
FP6
No verified data available No verified data available
FP4
No verified data available No verified data available
INT8
35.9 TFlops
71.8 TOPS sparse
284.0 TFlops
568.0 TOPS sparse

FLOPS by Precision

What actually differs

The A16 is the newer part: Ampere, launched in 2021, against the GeForce RTX 3090's Ampere from 2020. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.

Dense throughput for the A16 against the GeForce RTX 3090: FP32 4.5 vs 36 TFLOPS, FP16 18 vs 71 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 16 GB against 24 GB, fed at 200 GB/s versus 936 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.

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 A16 faster than the GeForce RTX 3090?

At FP16 precision the GeForce RTX 3090 reaches 71 TFLOPS dense against 18 TFLOPS for the A16. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the A16 or the GeForce RTX 3090?

The GeForce RTX 3090 carries 24 GB of memory versus 16 GB for the A16. Memory bandwidth is 200 GB/s for the A16 and 936 GB/s for the GeForce RTX 3090.

Can I rent the A16 or the GeForce RTX 3090 in the cloud?

Yes. Live cloud listings tracked by Flopper start at $0.11 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA GeForce RTX 3090 24GB

ProviderConfigurationPrice/GPU-hrChecked
Vast.ai
4× GeForce RTX 3090 24GB Community $0.11 5h ago View →
RunPod
1× GeForce RTX 3090 24GB Community $0.22 5h ago View →
RunPod
1× GeForce RTX 3090 24GB $0.50 5h ago View →
All GeForce RTX 3090 listings and price history →

Get Comparison Updates

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