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	309 fps 3.7 ms @ 30Hz	184 fps 18 ms @ 30Hz	108 fps 17 ms @ 30Hz	60.8 fps 19 ms @ 30Hz	413 fps 2.8 ms @ 30Hz	478 fps 2.2 ms @ 30Hz	233 fps 2.4 ms @ 30Hz
FoundationPose Pose Estimation Node	720p	2.21 fps 460 ms @ 30Hz	1.59 fps 760 ms @ 30Hz	0.502 fps 3800 ms @ 30Hz	–	1.64 fps 690 ms @ 30Hz	5.76 fps 170 ms @ 30Hz	3.11 fps 340 ms @ 30Hz
Stereo Disparity Node	1080p	187 fps 5.8 ms @ 30Hz	128 fps 27 ms @ 30Hz	86.5 fps 25 ms @ 30Hz	45.5 fps 22 ms @ 30Hz	198 fps 5.0 ms @ 30Hz	478 fps 2.4 ms @ 30Hz	281 fps 2.9 ms @ 30Hz
Rectify Node	1080p	1690 fps 0.48 ms @ 30Hz	1140 fps 0.52 ms @ 30Hz	886 fps 0.51 ms @ 30Hz	296 fps 1.4 ms @ 30Hz	2500 fps 0.15 ms @ 30Hz	2490 fps 0.43 ms @ 30Hz	1300 fps 0.46 ms @ 30Hz
TensorRT Node DOPE	VGA	192 fps 1.2 ms @ 30Hz	117 fps 2.2 ms @ 30Hz	31.1 fps 3.1 ms @ 30Hz	14.2 fps 3.0 ms @ 30Hz	122 fps 1.0 ms @ 30Hz	350 fps 0.78 ms @ 30Hz	46.8 fps 1.2 ms @ 30Hz
Triton Node DOPE	VGA	174 fps 5.9 ms @ 30Hz	104 fps 41 ms @ 30Hz	28.7 fps 52 ms @ 30Hz	13.0 fps 78 ms @ 30Hz	104 fps 8.9 ms @ 30Hz	310 fps 3.9 ms @ 30Hz	137 fps 7.3 ms @ 30Hz
TensorRT Node PeopleSemSegNet	544p	857 fps 0.96 ms @ 30Hz	376 fps 1.6 ms @ 30Hz	356 fps 1.9 ms @ 30Hz	184 fps 3.2 ms @ 30Hz	1110 fps 0.95 ms @ 30Hz	1520 fps 1.4 ms @ 30Hz	545 fps 1.5 ms @ 30Hz
Triton Node PeopleSemSegNet	544p	184 fps 5.3 ms @ 30Hz	99.7 fps 23 ms @ 30Hz	99.9 fps 13 ms @ 30Hz	83.4 fps 13 ms @ 30Hz	257 fps 4.6 ms @ 30Hz	1170 fps 1.6 ms @ 30Hz	412 fps 1.8 ms @ 30Hz
DNN Image Encoder Node	VGA	3100 fps 0.57 ms @ 30Hz	1890 fps 0.63 ms @ 30Hz	1010 fps 0.89 ms @ 30Hz	1090 fps 1.0 ms @ 30Hz	3030 fps 0.29 ms @ 30Hz	3110 fps 0.43 ms @ 30Hz	1170 fps 0.49 ms @ 30Hz
Occupancy Grid Localizer Node	~50 sq. m	1.28 fps 740 ms @ 30Hz	1.34 fps 770 ms @ 30Hz	0.711 fps 210 ms @ 30Hz	0.68 fps 840 ms @ 30Hz	2.11 fps 500 ms @ 30Hz	2.33 fps 460 ms @ 30Hz	2.56 fps 420 ms @ 30Hz
H.264 Decoder Node	1080p	713 fps 1.0 ms @ 30Hz	708 fps 3.0 ms @ 30Hz	260 fps 10 ms @ 30Hz	–	923 fps 1.3 ms @ 30Hz	1080 fps 0.82 ms @ 30Hz	1100 fps 0.82 ms @ 30Hz
H.264 Encoder Node I-frame Support	1080p	711 fps 2.6 ms @ 30Hz	708 fps 10 ms @ 30Hz	383 fps 16 ms @ 30Hz	–	927 fps 2.4 ms @ 30Hz	1090 fps 1.6 ms @ 30Hz	1130 fps 1.8 ms @ 30Hz
H.264 Encoder Node P-frame Support	1080p	728 fps 2.4 ms @ 30Hz	720 fps 11 ms @ 30Hz	442 fps 17 ms @ 30Hz	–	912 fps 2.0 ms @ 30Hz	1130 fps 1.6 ms @ 30Hz	1170 fps 1.9 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 4.3 ms @ 30Hz	187 fps 20 ms @ 30Hz	112 fps 19 ms @ 30Hz	59.0 fps 19 ms @ 30Hz	445 fps 3.9 ms @ 30Hz	520 fps 3.1 ms @ 30Hz	232 fps 5.9 ms @ 30Hz
Centerpose Pose Estimation Graph	VGA	50.2 fps 0.46 ms @ 30Hz	50.0 fps 0.81 ms @ 30Hz	45.1 fps 1.9 ms @ 30Hz	25.1 fps 1.4 ms @ 30Hz	50.2 fps 2.6 ms @ 30Hz	50.2 fps 0.42 ms @ 30Hz	50.2 fps 1.8 ms @ 30Hz
DNN Stereo Disparity Graph Full	576p	91.8 fps 12 ms @ 30Hz	61.4 fps 49 ms @ 30Hz	35.7 fps 52 ms @ 30Hz	–	97.9 fps 11 ms @ 30Hz	121 fps 9.6 ms @ 30Hz	54.1 fps 14 ms @ 30Hz
DNN Stereo Disparity Graph Light	288p	156 fps 9.1 ms @ 30Hz	102 fps 21 ms @ 30Hz	61.0 fps 28 ms @ 30Hz	–	178 fps 5.1 ms @ 30Hz	205 fps 6.7 ms @ 30Hz	84.7 fps 10 ms @ 30Hz
Stereo Disparity Graph	1080p	168 fps 6.1 ms @ 30Hz	115 fps 30 ms @ 30Hz	83.0 fps 25 ms @ 30Hz	41.2 fps 24 ms @ 30Hz	174 fps 6.0 ms @ 30Hz	413 fps 2.8 ms @ 30Hz	237 fps 3.2 ms @ 30Hz
DetectNet Object Detection Graph	544p	237 fps 4.9 ms @ 30Hz	153 fps 26 ms @ 30Hz	51.5 fps 44 ms @ 30Hz	21.7 fps 48 ms @ 30Hz	125 fps 8.7 ms @ 30Hz	341 fps 4.3 ms @ 30Hz	161 fps 6.8 ms @ 30Hz
Grounding DINO Object Detection Graph	544p	24.4 fps 45 ms @ 30Hz	15.3 fps 100 ms @ 30Hz	5.74 fps 210 ms @ 30Hz	–	17.3 fps 62 ms @ 30Hz	167 fps 10 ms @ 30Hz	67.5 fps 22 ms @ 30Hz
RT-DETR Object Detection Graph SyntheticaDETR	720p	220 fps 7.3 ms @ 30Hz	150 fps 30 ms @ 30Hz	61.9 fps 39 ms @ 30Hz	29.5 fps 38 ms @ 30Hz	186 fps 6.5 ms @ 30Hz	589 fps 4.4 ms @ 30Hz	215 fps 7.3 ms @ 30Hz
TensorRT Graph PeopleSemSegNet	544p	495 fps 4.3 ms @ 30Hz	266 fps 16 ms @ 30Hz	234 fps 11 ms @ 30Hz	99.6 fps 10 ms @ 30Hz	7.85 fps 49 ms @ 30Hz	796 fps 3.6 ms @ 30Hz	350 fps 4.5 ms @ 30Hz
SAM Image Segmentation Graph Full SAM	720p	4.98 fps 210 ms @ 30Hz	2.25 fps 380 ms @ 30Hz	1.31 fps 350 ms @ 30Hz	–	2.22 fps 270 ms @ 30Hz	20.8 fps 46 ms @ 30Hz	–
SAM Image Segmentation Graph Mobile SAM	720p	20.5 fps 51 ms @ 30Hz	13.6 fps 120 ms @ 30Hz	6.68 fps 170 ms @ 30Hz	–	14.6 fps 65 ms @ 30Hz	87.7 fps 14 ms @ 30Hz	47.4 fps 24 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) are measured using the maximum sustained framerate for the accelerated computational pipeline on each listed platforms. Performance measured includes input node → graph of nodes under performance test → output node where the publishing rate of the input node is auto-tuned to discover the peak throughput dropping <5% of the frames. Average fps computed over 5 runs, discarding minimum and maximum runs; latency measured at 30hz publishing rate matching sensor capture rate.