NVIDIA Jetson AGX Orin 64GB vs NVIDIA A10G 24GB

The A10G delivers 63% more FP16 throughput than the Jetson AGX Orin (70 vs 43 TFLOPS dense). The Jetson AGX Orin counters with 40 GB more memory (64 GB vs 24 GB).

Jetson AGX Orin: Ampere, 2021 A10G: Ampere, 2022
FP16 dense lead
+63%
A10G: 70 vs 43 TFLOPS
VRAM
64 vs 24 GB
40 GB more for the Jetson AGX Orin
Bandwidth
205 GB/s vs 600 GB/s
A10G moves data faster
TDP
60 W vs 300 W
Jetson AGX Orin draws 240 W less

Specifications

Jetson AGX Orin
A10G
Architecture
AmpereAmpere
Launch Year
20212022
Form Factor
ModulePCIe
VRAM
64 GB24 GB
Memory Bandwidth
205 GB/s600 GB/s
TDP
60 W300 W
Process Node
8nm

Performance (TFLOPS)

Jetson AGX Orin
A10G
FP64
No verified data available No verified data available
FP32
5.3 TFlops 35.0 TFlops
TF32
No verified data available35.0 TFlops
sparse not published
BF16
No verified data available70.0 TFlops
sparse not published
FP16
43.0 TFlops
85.0 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
85.0 TFlops
170.0 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 Jetson AGX Orin'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 Jetson AGX Orin against the A10G: FP32 5.3 vs 35 TFLOPS, FP16 43 vs 70 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 64 GB against 24 GB, fed at 205 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.

Power budgets are 60 W for the Jetson AGX Orin and 300 W for the A10G. At FP32 that works out to 0.09 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 Jetson AGX Orin faster than the A10G?

At FP16 precision the A10G reaches 70 TFLOPS dense against 43 TFLOPS for the Jetson AGX Orin. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the Jetson AGX Orin or the A10G?

The Jetson AGX Orin carries 64 GB of VRAM versus 24 GB for the A10G. Memory bandwidth is 205 GB/s for the Jetson AGX Orin and 600 GB/s for the A10G.

How much power do the Jetson AGX Orin and the A10G draw?

The Jetson AGX Orin is rated at 60 W TDP and the A10G at 300 W. On FP32 throughput per watt, the A10G is the more efficient part.

Get Comparison Updates

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