Home › Benchmarks › Jetson AGX Orin 64GB
Benchmark Explorer · Jetson AGX Orin 64GB
Jetson AGX Orin 64GB: 63 benchmark rows, 37% measured.
Method v1.0 · dataset 2026-09-07 · verified 2026-09-07 · last updated September 2026
23 of 63 rows are measured and published (class C); the rest are EdgeAIStack-derived (class D). 16 model families. Precisions: FP32 (10), FP16 (29), INT8 (22), INT4 (2). Runtimes: tensorrt (56), mlc (2), pytorch (5). Tasks: detection (48), classification (5), text generation (2), segmentation (3), lidar detection 3d (1), speech recognition (3), pose (1). Jetson AGX Orin 64GB 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.
| Family | FP32 | FP16 | INT8 | INT4 |
|---|---|---|---|---|
| EfficientDet | — | 310 fps D | 395.8 fps D | — |
| EfficientNet | — | 2800 fps D | — | — |
| Llama | — | — | — | 163.9 tok/s C |
| SSD MobileNet | — | — | 512 fps D | — |
| MobileSAM | 25.6 fps C | — | — | — |
| NanoSAM | — | 123.5 fps C | — | — |
| PointPillars | — | 25 fps D | — | — |
| ResNet | — | 2800 fps D | 1566 fps C | — |
| RetinaNet | — | 85.7 fps C | 85.7 fps C | — |
| RT-DETR | — | 101 fps D | 134 fps D | — |
| Whisper | — | — D | — | — |
| YOLO11 | 254.5 fps C | 392 fps C | 459 fps C | — |
| YOLO12 | — | 333 fps D | 416.5 fps D | — |
| YOLOv8 | — | 371 fps D | 504 fps D | — |
| YOLOv8 | — | 72 fps D | — | — |
| YOLOv8 | — | 95 fps D | — | — |
Every model on Jetson AGX Orin 64GB
EfficientDetd0
EfficientDet · d0
EfficientDetd1
EfficientDet · d1
EfficientDetd2
EfficientDet · d2
EfficientNetb0
EfficientNet · b0
EfficientNetb4
EfficientNet · b4
Llama 3.2 1B Instruct
Llama · llama-3.2-1b-instruct
Llama 3.2 3B Instruct
Llama · llama-3.2-3b-instruct
SSD MobileNetv1
SSD MobileNet · v1
MobileSAM Tiny Vit
MobileSAM · tiny-vit
NanoSAM Resnet18
NanoSAM · resnet18
PointPillarsbase
PointPillars · base
ResNet101
ResNet · 101
ResNet50
ResNet · 50
RetinaNetbase
RetinaNet · base
RetinaNet Resnet50
RetinaNet · resnet50
RT-DETRl
RT-DETR · l
Whisperbase
Whisper · base
Whisper Small
Whisper · small
Whispertiny
Whisper · tiny
YOLO11l
YOLO11 · l
YOLO11m
YOLO11 · m
YOLO11n
YOLO11 · n
YOLO11s
YOLO11 · s
YOLO11x
YOLO11 · x
YOLO12l
YOLO12 · l
YOLO12m
YOLO12 · m
YOLO12n
YOLO12 · n
YOLO12s
YOLO12 · s
YOLO12x
YOLO12 · x
YOLOv8l
YOLOv8 · l
YOLOv8m
YOLOv8 · m
YOLOv8n
YOLOv8 · n
YOLOv8s
YOLOv8 · s
YOLOv8x
YOLOv8 · x
YOLOv8m-pose
YOLOv8 · m-pose
YOLOv8m-seg
YOLOv8 · m-seg
Links
- Live engine: https://edgeaistack.ai/engines/benchmark-explorer/?hw=jetson_agx_orin&model=efficientdet&variant=d0&mv=1.0&dv=2026-09-07
- Methodology: /methodology/benchmark-explorer/
- Dataset: /datasets/benchmarks-v2.json
- Platform page: Jetson AGX Orin 64GB platform page
- All hardware: /benchmarks/
Query this hardware live.
Filter by model, precision or runtime and get the matching rows with a permanent link.