Back to browse
GitHub Repository

NumPy for the browser GPU — zero shaders, zero dependencies

5 starsTypeScript

Accel-GPU v1.0 – NumPy for browser GPU

by Phantasm0009·Mar 4, 2026·1 point·1 comment

AI Analysis

●●●BangerWizardryShip ItSolve My Problem

WebGPU NumPy alternative that beats TensorFlow.js on speed and simplicity.

Strengths
  • Shader-free WebGPU API eliminates WGSL boilerplate; built-in conv2d, FFT, matmul kernels
  • Universal fallback stack (WebGPU→WebGL2→CPU) ensures compatibility across Safari, Firefox, Node
  • 160KB minified, zero runtime dependencies; Arrow interop for data pipelines
Weaknesses
  • API surface narrower than TensorFlow.js; no pre-trained models or deployment tools
  • Limited community/ecosystem vs. established ML frameworks; early-stage maturity
Target Audience

Frontend developers, AI researchers, data engineers working in the browser

Similar To

TensorFlow.js · GPU.js · Danfo.js

Similar Projects

AI/ML●●●Banger

Auto GPU Kernel – Autonomous GPU-kernel discovery and optimizer

Autonomous kernel optimizer that won MLSys contest with 34.93x speedup.

WizardryBig BrainBold Bet
dogacel
109d ago