microsoft/onnxjs

Health v2 · maintenance only

Flags: archived no_license

Adoption not part of the score

1814 stars · 130 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

ONNX.js is a JavaScript library for running pre-trained ONNX machine learning models directly in browsers and Node.js, using WebAssembly for CPU and WebGL for GPU inference. It has been deprecated and replaced by ONNX Runtime Web, which offers better performance and an improved developer experience.

Use cases

• run ONNX models in the browser
• score machine learning models client-side with JavaScript
• run ML inference in Node.js
• do in-browser inference without a server round-trip
• run ONNX models on GPU via WebGL
• keep user data private by doing ML inference locally

When to choose

• maintaining a legacy project already built on onnxjs
• learning how ONNX inference runtimes work in JavaScript

When to avoid

• starting any new project - use ONNX Runtime Web instead
• needing the latest ONNX opset support or best performance
• needing actively maintained software with recent fixes

Facets

library · maturity abandoned

machine-learning llm-inference wasm machine-learning deep-learning web-development frontend browser wasm cross-platform onnx inference webgl webassembly deprecated typescript edge-ai nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="microsoft/onnxjs")