Description
Overview
The NVIDIA Jetson Orin Nano 4GB is the most cost-effective entry point into the Jetson Orin ecosystem. With 34 TOPS of AI performance powered by the NVIDIA Ampere architecture, it brings real AI inference capabilities to high-volume, cost-sensitive products where every dollar matters.
Despite its entry-level positioning, the Orin Nano 4GB features 512 CUDA cores, 16 Tensor Cores, and a 6-core Arm CPU — delivering more AI performance than entire previous-generation Jetson platforms. Its pin-compatible SO-DIMM form factor means you can design once and scale across the entire Orin Nano and Orin NX family.
Key Features
- 34 TOPS AI Performance: Real-time AI inference for cost-optimized edge deployments
- 512 CUDA Cores + 16 Tensor Cores: Ampere GPU acceleration at the lowest price point in the Orin family
- 4GB LPDDR5 Memory: 51 GB/s bandwidth for efficient single-model inference
- 6-Core Arm CPU: Cortex-A78AE at 1.7 GHz for balanced compute
- Ultra-Low Power: 7W to 25W for fanless and battery-powered designs
- Pin-Compatible: Same 260-pin SO-DIMM as Orin Nano 8GB and Orin NX modules
Technical Specifications
| Specification | Details |
|---|---|
| GPU Architecture | NVIDIA Ampere |
| CUDA Cores | 512 |
| Tensor Cores | 16 |
| GPU Frequency | 1.02 GHz |
| CPU | 6-core Arm Cortex-A78AE @ 1.7 GHz |
| Memory | 4GB 64-bit LPDDR5 |
| Memory Bandwidth | 51 GB/s |
| Storage | External NVMe via M.2 Key M |
| AI Performance | 34 TOPS (Sparse INT8) |
| Power | 7W – 25W (configurable) |
| Connector | 260-pin SO-DIMM |
| Module Size | 69.6 mm × 45 mm |
Ideal Use Cases
- High-volume smart cameras with on-device object detection
- Cost-optimized IoT edge gateways with AI classification
- Entry-level robotics and educational AI platforms
- Retail point-of-sale systems with computer vision
- Smart home and building automation with local AI
Why Choose This Product?
The Jetson Orin Nano 4GB makes edge AI economically viable for mass-market products. At 34 TOPS, it outperforms the entire previous-generation Jetson Nano platform while maintaining the same low-power, compact profile. For products shipping in thousands or tens of thousands of units, this module delivers the best AI-per-dollar in the NVIDIA embedded ecosystem.
Interested? Contact us for volume pricing, BOM optimization guidance, and production support.
