Performance Summary

The performance results of benchmarking the Isaac ROS packages on both Jetson and x86_64 platforms are below:

Node	Input Size	AGX Thor T5000	AGX Thor T4000	AGX Orin	Orin Nano Super 8GB	DGX Spark	x86_64 w/ RTX 5090	x86_64 w/ RTX 5070
AprilTag Node	720p	326 fps 3.2 ms @ 30Hz	248 fps 4.3 ms @ 30Hz	189 fps 5.3 ms @ 30Hz	104 fps 9.6 ms @ 30Hz	555 fps 1.5 ms @ 30Hz	596 fps 1.0 ms @ 30Hz	591 fps 1.4 ms @ 30Hz
FoundationPose Pose Estimation Node	720p	2.42 fps	1.82 fps	0.746 fps	–	1.67 fps	5.16 fps	3.32 fps
Stereo Disparity Node	1080p	171 fps 5.9 ms @ 30Hz	141 fps 8.4 ms @ 30Hz	124 fps 8.6 ms @ 30Hz	66.1 fps 17 ms @ 30Hz	156 fps 4.9 ms @ 30Hz	512 fps 2.1 ms @ 30Hz	415 fps 2.5 ms @ 30Hz
Rectify Node	1080p	1150 fps 0.27 ms @ 30Hz	683 fps 0.27 ms @ 30Hz	1220 fps 0.28 ms @ 30Hz	470 fps 0.58 ms @ 30Hz	2030 fps 0.24 ms @ 30Hz	7500 fps 0.12 ms @ 30Hz	5680 fps 0.13 ms @ 30Hz
TensorRT Node DOPE	VGA	193 fps 1.4 ms @ 30Hz	128 fps 1.3 ms @ 30Hz	47.4 fps 1.9 ms @ 30Hz	20.8 fps	108 fps 1.2 ms @ 30Hz	350 fps 0.53 ms @ 30Hz	161 fps 0.48 ms @ 30Hz
Triton Node DOPE	VGA	174 fps 6.2 ms @ 30Hz	116 fps 10 ms @ 30Hz	45.2 fps 22 ms @ 30Hz	20.2 fps	98.1 fps 8.6 ms @ 30Hz	350 fps 2.9 ms @ 30Hz	156 fps 6.4 ms @ 30Hz
TensorRT Node PeopleSemSegNet	544p	792 fps 1.2 ms @ 30Hz	524 fps 1.4 ms @ 30Hz	481 fps 1.5 ms @ 30Hz	234 fps 1.8 ms @ 30Hz	1030 fps 0.87 ms @ 30Hz	3000 fps 0.46 ms @ 30Hz	2320 fps 0.52 ms @ 30Hz
Triton Node PeopleSemSegNet	544p	195 fps 6.3 ms @ 30Hz	191 fps 6.7 ms @ 30Hz	189 fps 5.5 ms @ 30Hz	122 fps 8.4 ms @ 30Hz	173 fps 5.7 ms @ 30Hz	2440 fps 0.53 ms @ 30Hz	1630 fps 0.74 ms @ 30Hz
DNN Image Encoder Node	1200p	767 fps 2.5 ms @ 30Hz	418 fps 4.6 ms @ 30Hz	423 fps 4.5 ms @ 30Hz	475 fps 4.5 ms @ 30Hz	1130 fps 1.9 ms @ 30Hz	2870 fps 0.91 ms @ 30Hz	449 fps 2.5 ms @ 30Hz
Occupancy Grid Localizer Node	~50 sq. m	1.46 fps	1.42 fps	1.16 fps	1.05 fps	2.39 fps	2.64 fps	2.42 fps
H.264 Decoder Node	1200p	500 fps 2.4 ms @ 30Hz	513 fps 1.9 ms @ 30Hz	262 fps 6.6 ms @ 30Hz	162 fps 9.7 ms @ 30Hz	313 fps 1.4 ms @ 30Hz	1050 fps 0.86 ms @ 30Hz	839 fps 1.1 ms @ 30Hz
H.264 Encoder Node I-frame Support	1080p	766 fps 2.2 ms @ 30Hz	777 fps 2.2 ms @ 30Hz	399 fps 15 ms @ 30Hz	–	955 fps 1.6 ms @ 30Hz	1110 fps 1.7 ms @ 30Hz	1120 fps 1.9 ms @ 30Hz
H.264 Encoder Node P-frame Support	1080p	779 fps 2.5 ms @ 30Hz	492 fps 2.2 ms @ 30Hz	463 fps 11 ms @ 30Hz	–	927 fps 1.4 ms @ 30Hz	1150 fps 2.0 ms @ 30Hz	1150 fps 2.0 ms @ 30Hz

Graph	Input Size	AGX Thor T5000	AGX Thor T4000	AGX Orin	Orin Nano Super 8GB	DGX Spark	x86_64 w/ RTX 5090	x86_64 w/ RTX 5070
AprilTag Graph	720p	306 fps 3.6 ms @ 30Hz	236 fps 5.0 ms @ 30Hz	186 fps 6.5 ms @ 30Hz	100 fps 12 ms @ 30Hz	539 fps 2.3 ms @ 30Hz	596 fps 1.7 ms @ 30Hz	596 fps 2.3 ms @ 30Hz
Centerpose Pose Estimation Graph	VGA	50.2 fps 1.8 ms @ 30Hz	50.2 fps 1.9 ms @ 30Hz	50.2 fps 1.4 ms @ 30Hz	35.0 fps 1.8 ms @ 30Hz	50.2 fps 0.58 ms @ 30Hz	50.2 fps 0.21 ms @ 30Hz	50.2 fps 0.25 ms @ 30Hz
DNN Stereo Disparity Graph Full	576p	99.8 fps 12 ms @ 30Hz	77.1 fps 16 ms @ 30Hz	66.5 fps 17 ms @ 30Hz	34.8 fps 30 ms @ 30Hz	102 fps 10 ms @ 30Hz	350 fps 3.2 ms @ 30Hz	233 fps 4.7 ms @ 30Hz
DNN Stereo Disparity Graph Light	288p	203 fps 6.7 ms @ 30Hz	161 fps 8.3 ms @ 30Hz	75.5 fps 24 ms @ 30Hz	82.2 fps 13 ms @ 30Hz	192 fps 5.4 ms @ 30Hz	350 fps 2.5 ms @ 30Hz	350 fps 3.0 ms @ 30Hz
Stereo Disparity Graph	1080p	159 fps 6.2 ms @ 30Hz	129 fps 9.2 ms @ 30Hz	117 fps 9.3 ms @ 30Hz	61.5 fps 19 ms @ 30Hz	143 fps 5.3 ms @ 30Hz	450 fps 2.2 ms @ 30Hz	380 fps 2.6 ms @ 30Hz
DetectNet Object Detection Graph	544p	216 fps 7.7 ms @ 30Hz	157 fps 8.4 ms @ 30Hz	73.5 fps 15 ms @ 30Hz	30.4 fps 37 ms @ 30Hz	120 fps 8.9 ms @ 30Hz	291 fps 4.7 ms @ 30Hz	166 fps 7.1 ms @ 30Hz
Grounding DINO Object Detection Graph	544p	25.3 fps	17.5 fps	13.6 fps	–	17.3 fps	156 fps 8.0 ms @ 30Hz	65.6 fps 17 ms @ 30Hz
RT-DETR Object Detection Graph SyntheticaDETR	720p	195 fps 6.7 ms @ 30Hz	177 fps 8.3 ms @ 30Hz	87.3 fps 13 ms @ 30Hz	40.0 fps 27 ms @ 30Hz	181 fps 6.8 ms @ 30Hz	718 fps 2.3 ms @ 30Hz	374 fps 3.6 ms @ 30Hz
TensorRT Graph PeopleSemSegNet	544p	478 fps 3.8 ms @ 30Hz	352 fps 4.4 ms @ 30Hz	315 fps 4.7 ms @ 30Hz	156 fps 8.0 ms @ 30Hz	298 fps 2.9 ms @ 30Hz	1160 fps 1.1 ms @ 30Hz	935 fps 1.6 ms @ 30Hz
SAM Image Segmentation Graph Full SAM	720p	5.14 fps	2.22 fps	2.22 fps	–	2.22 fps	25.0 fps	8.40 fps
SAM Image Segmentation Graph Mobile SAM	720p	20.8 fps	14.6 fps	8.40 fps	4.80 fps	14.6 fps	109 fps 10 ms @ 30Hz	57.9 fps 18 ms @ 30Hz

Methodology

The performance numbers provided above are measured using Isaac ROS Benchmark. Configuration files for reproducing results are provided in the launch scripts for each benchmark here.

Frames per second (fps) report observed throughput from the input node through the computational pipeline to the output node. The input rate is auto-tuned with a target of less than 5% frame loss; these approximate results include some runs slightly above that target. Average fps is computed over five runs, discarding the minimum and maximum. Latency is the time from the first sent frame to the first received frame in a separate 30 Hz trial from the same benchmark run. Latency is omitted when no valid 30 Hz measurement is available.