OminiX Edge
High-performance Inference

Optimized inference engines for consumer hardware and edge devices. Run state-of-the-art models efficiently without the cloud tax.

Explore Compute Resources

View our open source repositories and documentation.

github.com/moxin-org

OminiX Edge

Bring intelligence to the edge. Run optimized models on iPhone, MacBook, and NPU-enabled devices with minimal battery drain.

  • Metal & CoreML Optimization
  • Local-First Privacy
  • Zero-Latency Offline Mode