Description
Overview
The NVIDIA IGX Thor Developer Kit Mini brings Blackwell-class industrial AI to a compact form factor. With the integrated 2,560-core Blackwell GPU delivering 2,070 TFLOPS of FP4 sparse performance and 128GB of LPDDR5X memory, it provides a powerful yet space-efficient development platform for functional safety-critical edge AI applications.
Key Features
- 2,070 TFLOPS FP4 Sparse: Blackwell iGPU performance without a discrete GPU
- 128GB LPDDR5X: Large unified memory for complex AI models
- Compact Design: 243 x 112 x 57 mm — fits space-constrained deployments
- Wi-Fi 6E + 5GbE: Flexible connectivity options
- Functional Safety: ISO 26262 ASIL D / IEC 61508 SIL 2 capable
Technical Specifications
| Specification | Details |
|---|---|
| GPU | 2,560-core NVIDIA Blackwell iGPU |
| CPU | 14-core Arm Neoverse-V3AE @ 2.6 GHz |
| Memory | 128 GB LPDDR5X |
| AI Performance | 2,070 TFLOPS (FP4 Sparse) |
| Storage | 1 TB NVMe + UFS3.1 |
| Networking | 1x 5GbE, 1x QSFP28, Wi-Fi 6E |
| Power | 40–130W |
| Safety | ISO 26262 ASIL D / IEC 61508 SIL 2 |
| Dimensions | 243 x 112 x 57 mm |
Ideal Use Cases
- Compact medical AI devices — point-of-care diagnostics, portable imaging
- Space-constrained industrial AI controllers
- Prototyping safety-critical AI without discrete GPU requirements
- Edge AI gateways for smart factory deployments
Why Choose This Product?
When you need functional safety-certified AI compute in the smallest possible package, the IGX Thor Mini delivers 2,070 TFLOPS without a discrete GPU — simplifying thermal design, reducing BOM cost, and fitting into enclosures where the full developer kit cannot.
Interested? Contact us for availability, integration support, and safety certification guidance.
