Experience the world’s most powerful AI computer in the smallest Jetson form factor for power-efficient autonomous machines. It delivers 5x more performance compared to the previous generation. With 100 TOPS for simultaneous multi-AI inference streams, twice the performance from CUDA cores, and high-speed interface support for multiple sensors, the Jetson Orin NX delivers massive performance in an incredibly compact package.
Jetson Orin NX 8GB
| AI Performance | 70 TOPS (INT8) |
|---|---|
| GPU | NVIDIA Ampere architecture with 1024 NVIDIA® CUDA® cores and 32 tensor cores |
| Max GPU Freq | 765 MHz |
| CPU | 6-core Arm® Cortex®-A78AE v8.2 64-bit CPU 1.5MB L2 + 4MB L3 |
| CPU Max Freq | 2 GHz |
| DL Accelerator | 1x NVDLA v2.0 |
| DLA Max Freq | 614 MHz |
| Vision Accelerator | PVA v2.0 |
| Memory | 8GB 128-bit LPDDR5 102.4 GB/s |
| Storage | Supports external NVMe |
| CSI Camera | Up to 4 cameras (8 via virtual channels*) 8 MIPI CSI-2 lanes D-PHY 1.2 (20 Gbps) |
| Video Encode | 1x 4K60 | 3x 4K30| 6x 1080p60 | 12x 1080p30 (H.265) H.264, H.265, AV1 |
| Video Decode | 1x 8K30 | 2x 4K60 | 4x 4K30 | 9x 1080p60| 18x 1080p30 (H.265) H.264, VP9, AV1 |
| UPHY | 3 x1 + 1 x4 PCIe Gen 4 3x USB 3.2 Gen2 |
| Networking | 1x GbE |
| Display | 1x 8K30 multi-mode DP 1.4a (+MST)/eDP 1.4a/HDMI 2.1 |
| Other I/O | 3x USB 2.0 3x UART | 2x SPI | 4x I2C | 1x CAN | DMIC | DSPK | 2x I2S | 15x GPIOs |
| Power | 10W – 20W |
| Mechanical | 69.6mm x 45mm 260-pin SO-DIMM connector |
Jetson Orin NX 16GB
| GPU | NVIDIA Ampere architecture with 1024 NVIDIA® CUDA® cores and 32 tensor cores |
|---|---|
| CPU | 8-core Arm® Cortex®-A78AE v8.2 64-bit CPU 2MB L2 + 4MB L3 |
| DL Accelerator | 2x NVDLA v2.0 |
| DLA Max Freq | 614 MHz |
| Vision Accelerator | PVA v2.0 |
| DL Accelerator | 1x NVDLA v2.0 |
| DLA Max Freq | 614 MHz |
| Vision Accelerator | PVA v2.0 |
| Memory | 16GB 128-bit LPDDR5 102.4 GB/s |
| Storage | Supports external NVMe |
| CSI Camera | Up to 4 cameras (8 via virtual channels*) 8 MIPI CSI-2 lanes D-PHY 1.2 (20 Gbps) |
| Video Encode | 1x 4K60 | 3x 4K30| 6x 1080p60 | 12x 1080p30 (H.265) H.264, H.265, AV1 |
| Video Decode | 1x 8K30 | 2x 4K60 | 4x 4K30 | 9x 1080p60| 18x 1080p30 (H.265) H.264, VP9, AV1 |
| UPHY | 3 x1 + 1 x4 PCIe Gen 4 3x USB 3.2 Gen2 |
| Networking | 1x GbE |
| Display | 1x 8K30 multi-mode DP 1.4a (+MST)/eDP 1.4a/HDMI 2.1 |
| Other I/O | 3x USB 2.0 3x UART | 2x SPI | 4x I2C | 1x CAN | DMIC | DSPK | 2x I2S | 15x GPIOs |
| Power | 10W – 25W |
| Mechanical | 69.6mm x 45mm 260-pin SO-DIMM connector |
Pre-trained Models
Many AI applications share common needs such as classification, object detection, language translation, text reading, recommendation engines, sentiment analysis, and more. Pre-trained, optimized models in the NGC catalog can be fine-tuned using the NVIDIA TAO Toolkit and users’ own datasets. This reduces the time and cost required to bring AI models into production.
