NVIDIA A16 16GB vs NVIDIA A10G 24GB

The A10G delivers 3.9x the FP16 throughput of the A16 (70 vs 18 TFLOPS dense). The A10G also carries 8 GB more memory (24 GB vs 16 GB).

A16: Ampere, 2021 A10G: Ampere, 2022
FP16 dense lead
3.9x
A10G: 70 vs 18 TFLOPS
Memory
16 vs 24 GB
8 GB more for the A10G
Bandwidth
200 GB/s vs 600 GB/s
A10G moves data faster

Specifications

A16
A10G
Architecture
AmpereAmpere
Launch Year
20212022
Form Factor
PCIePCIe
Memory
16 GB24 GB
Memory Bandwidth
200 GB/s600 GB/s
TDP
300 W
Process Node
8nm

Performance (TFLOPS)

A16
A10G
FP64
No verified data available No verified data available
FP32
4.5 TFlops 35.0 TFlops
TF32
9.0 TFlops
18.0 TFLOPS sparse
35.0 TFlops
sparse not published
BF16
No verified data available70.0 TFlops
sparse not published
FP16
17.9 TFlops
35.9 TFLOPS sparse
70.0 TFlops
sparse not published
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
140.0 TFlops
sparse not published

FLOPS by Precision

What actually differs

The A10G is the newer part: Ampere, launched in 2022, against the A16's Ampere from 2021. 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 A10G: FP32 4.5 vs 35 TFLOPS, FP16 18 vs 70 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 600 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 A10G?

At FP16 precision the A10G reaches 70 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 A10G?

The A10G carries 24 GB of memory versus 16 GB for the A16. Memory bandwidth is 200 GB/s for the A16 and 600 GB/s for the A10G.

Get Comparison Updates

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