NVIDIA Jetson AGX Thor T5000 128GB vs NVIDIA DGX Spark

Side-by-side specifications, performance, and rental pricing for datacenter AI workloads.

Jetson AGX Thor: Blackwell, 2025 DGX Spark: Grace Blackwell, 2025
VRAM
128 vs 128 GB
Same capacity
Bandwidth
273 GB/s vs 273 GB/s
Even
TDP
130 W vs 140 W
Jetson AGX Thor draws 10 W less

Specifications

Jetson AGX Thor
DGX Spark
Architecture
BlackwellGrace Blackwell
Launch Year
20252025
Form Factor
ModuleDesktop Workstation
VRAM
128 GB128 GB
Memory Bandwidth
273 GB/s273 GB/s
TDP
130 W140 W
Process Node
4nm

Performance (TFLOPS)

Jetson AGX Thor
DGX Spark
FP64
No verified data available No verified data available
FP32
8.1 TFlops No verified data available
TF32
No verified data available No verified data available
BF16
No verified data available No verified data available
FP16
No verified data available No verified data available
FP8
517.0 TFlops
1035.0 TFLOPS sparse
No verified data available
FP6
No verified data available No verified data available
FP4
1035.0 TFlops
2070.0 TFLOPS sparse
500.0 TFlops
1000.0 TFLOPS sparse
INT8
517.0 TFlops
1035.0 TOPS sparse
No verified data available

FLOPS by Precision

What actually differs

Both GPUs launched in 2025: the Jetson AGX Thor on Blackwell and the DGX Spark on Grace Blackwell.

Memory is 128 GB against 128 GB, fed at 273 GB/s versus 273 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

Which has more memory, the Jetson AGX Thor or the DGX Spark?

Both GPUs carry 128 GB of VRAM. Memory bandwidth is 273 GB/s for the Jetson AGX Thor and 273 GB/s for the DGX Spark.

How much power do the Jetson AGX Thor and the DGX Spark draw?

The Jetson AGX Thor is rated at 130 W TDP and the DGX Spark at 140 W. On FP32 throughput per watt, the Jetson AGX Thor is the more efficient part.

Get Comparison Updates

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