The "Super" is a software re-rating, not new silicon. JetPack 6.2 raised the GPU clock ceiling from 625 MHz to 1020 MHz, the CPU from 1.5 GHz to 1.7 GHz and memory bandwidth from 68 GB/s to 102 GB/s on hardware already in the field, taking AI performance from 40 to 67 sparse INT8 TOPS while the developer kit price fell from $499 to $249. This row records the Super figures. POWER IS CONFIGURABLE and NVIDIA publishes three modes, 7W, 15W and 25W; tdp_watts and max_power_watts both carry the top mode, and reading either as a fixed rating would be wrong. NVIDIA publishes dense and sparse INT8 as two explicitly labelled rows for this part, so the dense figure needed no derivation. It publishes no FP16 and no FP32 rate for the Orin Nano, unlike the AGX Orin, and neither is derived here.

NVIDIA Jetson Orin Nano Super 8GB

Ampere Module 2024 Spec confidence: Official
VRAM
8 GB
LPDDR5
Bandwidth
102 GB/s
memory
TDP
25 W
Download datasheet

Overview

The Jetson Orin Nano Super Developer Kit is the cheapest way to get a modern NVIDIA tensor-core GPU with a supported CUDA and TensorRT stack, at $249. It is an 8 GB module with a 1024-core Ampere GPU and 32 third-generation tensor cores, giving 33 dense INT8 TOPS or 67 with structured sparsity, inside a configurable 7 W to 25 W envelope. The "Super" name refers to a JetPack 6.2 software update in December 2024 that lifted clocks and memory bandwidth on existing hardware and roughly two-thirds again the AI performance, alongside a halving of the price. What constrains it is memory: 8 GB of 128-bit LPDDR5 at 102 GB/s rules out anything but small quantised models, so this is a robotics, vision and edge-inference part rather than a machine for running a large language model. For learning the NVIDIA embedded stack it has no real competition on price.

Performance

Peak theoretical throughput by precision type

PrecisionDense2:4 Sparse
FP64
No verified data available Structured sparsity is a tensor-core feature; this vector precision has no sparse form
FP32
No verified data available Structured sparsity is a tensor-core feature; this vector precision has no sparse form
TF32
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
BF16
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP16
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP8
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP6
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP4
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
INT8
8-bit integer
33TOPS 67TOPS

Specifications

Architecture

Ampere

Form Factor

Module

Launch Year

2024

Process Node

No verified data available

Memory

8 GB LPDDR5

Bandwidth

102 GB/s

TDP

25 W

Max Power

25 W

CUDA Cores

1,024

Spec Confidence

Official

Full Specifications

Compute Engine
CUDA Cores 1,024
Tensor Cores 32 (3rd Gen)
Streaming Multiprocessors 8
Boost Clock 1.02 GHz
Memory
VRAM 8 GB
Memory Type LPDDR5
Bandwidth 102 GB/s
Interface Width 128-bit
Power & Thermal
TDP 25 W
Max Board Power 25 W
Enterprise Features
Sparsity Yes
General
Form Factor Module
Architecture Ampere
Process Node No verified data available
Launch Year 2024

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
2024-12-17

Data Quality

Spec confidence
Official
Clock basis
Boost
Core precisions with figures
1 of 9
Normalization
All values in TFLOPS

Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.

Browse GPU Cloud Providers

Similar GPUs

NVIDIA RTX A6000 48GB

PCIe · 2020
FP32: 39 TFLOPS
Compare vs RTX A6000

NVIDIA A40 48GB

PCIe · 2021
FP32: 37 TFLOPS
Compare vs A40

NVIDIA A10G 24GB

PCIe · 2022
FP32: 35 TFLOPS
Compare vs A10G

Frequently Asked Questions

How many TFLOPS does the NVIDIA Jetson Orin Nano Super have?

The verified peak figures Flopper holds for the NVIDIA Jetson Orin Nano Super are 33 TOPS INT8. Flopper does not currently have verified FP32, FP16 and FP8 throughput figures for it.

What is the power consumption of the NVIDIA Jetson Orin Nano Super?

The NVIDIA Jetson Orin Nano Super has a TDP (Thermal Design Power) rating of 25 watts.

How much memory does the NVIDIA Jetson Orin Nano Super have?

The NVIDIA Jetson Orin Nano Super is equipped with 8 GB of VRAM with 102 GB/s of memory bandwidth.

What architecture is the NVIDIA Jetson Orin Nano Super based on?

The NVIDIA Jetson Orin Nano Super is based on the Ampere architecture, launched in 2024.

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs