Jetson AGX Orin vs DGX Spark: how do they compare?
NVIDIA DGX Spark is faster in 4 of 4 like-for-like comparisons. Example, object detection (nano (YOLO-n), batch 1): NVIDIA DGX Spark 990 vs NVIDIA Jetson AGX Orin 64GB 435 FPS.
| Workload · group | NVIDIA Jetson AGX Orin 64GB | NVIDIA DGX Spark | Faster |
|---|---|---|---|
| object detection nano (YOLO-n) · batch 1 | 435 FPS B undated YOLO26n · source | 990 FPS B undated YOLO11n · source | NVIDIA DGX Spark (2.28×) |
| LLM generation LLM 5–14B · batch 1 | 30.00 tok/s C undated Qwen 3 8B · source | 60.39 tok/s C 2026-10-10 Llama 2 7B · source | NVIDIA DGX Spark (2.01×) |
| LLM generation unknown · batch 1 | 44.00 tok/s C undated Cosmos3 Edge (Reasoner) · source | 59.60 tok/s C undated Nemotron 3 Nano Omni · source | NVIDIA DGX Spark (1.35×) |
| LLM generation unknown · batch 8 | 171 tok/s C undated Cosmos3 Edge (Reasoner) · source | 191 tok/s C undated Nemotron 3 Nano Omni · source | NVIDIA DGX Spark (1.12×) |
Sources
- https://docs.ultralytics.com/guides/nvidia-dgx-spark/
- https://docs.ultralytics.com/guides/nvidia-jetson/
- https://github.com/ggml-org/llama.cpp/discussions/10879
- https://github.com/NVIDIA-AI-IOT/jetson-ai-lab/blob/main/src/data/benchmarks.json
Generated 2026-10-16 from HWBench data only. All answers · Submit a result
Machine-readable: /find.json (see llms.txt); full tables are in “Show data”. Confidence grades: A audited MLPerf closed · B documented third-party · C community · D vendor-published · E estimate. Unknown values are shown as unknown, never estimated. Generated 2026-10-16 from public sources.