Skip to content
Kitbay
Submit

transformers-js

Apache-2.0Updated on Nov 11, 202611,334GitHub stars

A Hugging Face skill on using the Transformers.js library to run machine learning models from JavaScript or TypeScript, in the browser or in Node.js, Bun and Deno, with WebGPU or WebAssembly.

What the skill teaches

A Hugging Face skill on using the Transformers.js library to run machine learning models from JavaScript or TypeScript, in the browser or in Node.js, Bun and Deno, with WebGPU or WebAssembly. It covers installing the package, the pipeline API, the supported text, vision, audio and multimodal tasks, finding suitable models on the Hub, configuration such as device and data type, differences between runtimes, error handling, performance, memory management and troubleshooting, with a quick reference of task ids and further reference files. Models download from the Hub unless local copies are used.

Capabilities

  • Run Hub models in JavaScript with Transformers.js
  • Use pipelines for text, vision and audio tasks
  • Configure WebGPU or WASM execution
  • Manage model memory in browser and server runtimes