GRX brings PyTorch-style tensor operations to Ruby. Every arithmetic op, activation, and optimizer step runs through a native C library with dynamic multi-target SIMD dispatch (AVX2+FMA, SSE, and scalar fallback). Ruby is the interface — C does the work. Features: autograd, SGD/Adam optimizers, Linear/Sequential/Dropout/BatchNorm layers, MSE/BCE/CrossEntropy loss functions, Xavier and He weight init. Cross-platform: .so on Linux, .dylib on macOS, .dll on Windows.

Required Ruby Version

>= 3.0.0

Authors

Razo

Versions

  1. 0.2.1

    1. source

      1. 0.2.1 source August 30, 2026 (70.5 KB)
  2. 0.2.0

    1. source

      1. 0.2.0 source August 22, 2026 (39.5 KB)
  3. 0.1.0

    1. source

      1. 0.1.0 source May 12, 2026 (27.5 KB)

Pushed by

GR

SHA 256 checksum

← Previous version