AI Potluck
Infrastructure / Edge AI hardware

NVIDIA Jetson Orin Nano Super Developer Kit

NVIDIA

An entry-level edge-AI developer kit pairing the Jetson Orin Nano 8GB module with a carrier board; the 'Super' mode boosts it to 67 INT8 (sparse) TOPS. It uses an Ampere GPU with 1024 CUDA cores and 32 Tensor Cores, a 6-core Arm Cortex-A78AE CPU, and 8GB LPDDR5. Made by NVIDIA, it was repriced to $249 and is the cheapest on-ramp into the CUDA/Jetson generative-AI ecosystem, capable of running small LLMs and VLMs locally.

Verified live 2026-06-22 via primary sources. Public datasheets and globally purchasable, but the SDK requires proprietary NVIDIA drivers/blobs under license, so it sits in the documented tier despite the community OE4T BSP.

Openness

3 high confidence
3.0
schematics
partial (carrier design guide public)
toolchain
closed (JetPack/Jetson Linux free but NVIDIA-SLA proprietary)
datasheets
public
blobs
required (proprietary GPU drivers/bootloader required)
retail
open_market (mass-market)

Public datasheets and globally purchasable, but the SDK requires proprietary NVIDIA drivers/blobs under license, so it sits in the documented tier despite the community OE4T BSP.

Adoption

4 high confidence
4.0

Widely available developer kit at a low $249 price with a large NVIDIA Jetson developer community and global distribution.

Capability

3 high confidence
3.0

Solid entry-tier generative-AI compute, but 8GB RAM caps model size versus higher Orin tiers.

Unchanged since 2026-08-01 (last edited, not re-checked)