Jetson AGX Orin Series
Bring your next-gen products to life with the world’s most powerful AI computer for energy-efficient autonomous machines, with up to 275 TOPS. 8x the performance of the last generation for multiple concurrent AI inference pipelines and high-speed interface support for multiple sensors make this the ideal solution for applications from manufacturing and logistics to retail and healthcare.
Technical specifications
| Specification | Jetson AGX Orin 64GB | Jetson AGX Orin 32GB | Jetson AGX Orin Industrial |
|---|---|---|---|
| AI performance | 275 TOPS | 241 TOPS | 248 TOPS |
| GPU | 2048-core NVIDIA Ampere GPU with 64 Tensor Cores | 1792-core NVIDIA Ampere GPU with 56 Tensor Cores | 2048-core NVIDIA Ampere GPU with 64 Tensor Cores |
| CPU | 12-core Arm Cortex-A78AE v8.2 64-bit, 3 MB L2 + 6 MB L3 | 8-core Arm Cortex-A78AE v8.2 64-bit, 2 MB L2 + 4 MB L3 | 12-core Arm Cortex-A78AE v8.2 64-bit, 3 MB L2 + 6 MB L3 |
| Memory | 64 GB 256-bit LPDDR5, 204.8 GB/s | 32 GB 256-bit LPDDR5, 204.8 GB/s | 64 GB 256-bit LPDDR5 with ECC, 204.8 GB/s |
| Storage | 64 GB eMMC 5.1 | 64 GB eMMC 5.1 | 64 GB eMMC 5.1 |
| Power | 15 W – 60 W | 15 W – 60 W | 15 W – 75 W |
| Operating temperature (at TTP surface) | -25 °C to 80 °C | -25 °C to 80 °C | -40 °C to 85 °C |
| Connector | 699-pin Molex Mirror Mezz, integrated thermal transfer plate | 699-pin Molex Mirror Mezz, integrated thermal transfer plate | 699-pin Molex Mirror Mezz, integrated thermal transfer plate |
| Dimensions | 100 mm x 87 mm | 100 mm x 87 mm | 100 mm x 87 mm |
For the 32 GB module, the compute comparison on the same NVIDIA page lists 200 sparse INT8 TOPS instead of 241.
Source: NVIDIA – Jetson Orin