HomeBenchmarks › Jetson Orin Nano Super
Benchmark Explorer · Jetson Orin Nano Super

Jetson Orin Nano Super: 58 benchmark rows, 40% measured.

Method v1.0 · dataset 2026-09-07 · verified 2026-09-07 · last updated September 2026

23 of 58 rows are measured and published (class C); the rest are EdgeAIStack-derived (class D). 18 model families. Precisions: FP32 (5), FP16 (25), INT8 (20), UNSPECIFIED (8). Runtimes: tensorrt (50), unspecified (8). Tasks: classification (9), detection (41), text generation (4), vision language (4). Jetson Orin Nano Super platform page.

Family × precision: best value

The fastest matching row per family and precision (fps, or tokens/s for language models), with its evidence class. Click a cell for the full breakdown.

FamilyFP32FP16INT8UNSPECIFIED
CLIP314 fps C
DINOv2126 fps C
EfficientDet162 fps D
Gemma 234.97 tok/s C
Llama43.07 tok/s C
LLaVA0.57 tok/s C
SSD MobileNet215 fps D
MobileNetV2740 fps D1090 fps D
MobileNetV3820 fps D
Qwen2.521.75 tok/s C
ResNet650 fps D
RT-DETR19.8 fps D28.2 fps D
SmolVLM12.9 tok/s C
VILA1.06 tok/s C
ViT273 fps C
YOLO1152 fps D221 fps C270 fps C
YOLO12187 fps D228.1 fps D
YOLOv869.1 fps D259 fps D341 fps D

Every model on Jetson Orin Nano Super

Query this hardware live.

Filter by model, precision or runtime and get the matching rows with a permanent link.

OPEN BENCHMARK EXPLORER →