\n
![\"transformers.js](\"https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/transformersjs-light.svg\")
\n \n
\n
\n \n
\n
\n ![\"NPM\"](\"https://img.shields.io/npm/v/@huggingface/transformers\")
![\"NPM](\"https://img.shields.io/npm/dw/@huggingface/transformers\")
![\"jsDelivr](\"https://img.shields.io/jsdelivr/npm/hw/@huggingface/transformers\")
![\"License\"](\"https://img.shields.io/github/license/huggingface/transformers.js?color=blue\")
![\"Documentation\"](\"https://img.shields.io/website/http/huggingface.co/docs/transformers.js/index.svg?down_color=red&down_message=offline&up_message=online\")
\n
State-of-the-art Machine Learning for the Web
\n\n\nRun đ€ Transformers directly in your browser, with no need for a server!\n\nTransformers.js is designed to be functionally equivalent to Hugging Face's [transformers](https://github.com/huggingface/transformers) python library, meaning you can run the same pretrained models using a very similar API. These models support common tasks in different modalities, such as:\n - đ **Natural Language Processing**: text classification, named entity recognition, question answering, language modeling, summarization, translation, multiple choice, and text generation.\n - đŒïž **Computer Vision**: image classification, object detection, segmentation, and depth estimation.\n - đŁïž **Audio**: automatic speech recognition, audio classification, and text-to-speech.\n - đ **Multimodal**: embeddings, zero-shot audio classification, zero-shot image classification, and zero-shot object detection.\n\nTransformers.js uses [ONNX Runtime](https://onnxruntime.ai/) to run models in the browser. The best part about it, is that you can easily [convert](#convert-your-models-to-onnx) your pretrained PyTorch, TensorFlow, or JAX models to ONNX using [đ€ Optimum](https://github.com/huggingface/optimum#onnx--onnx-runtime). \n\nFor more information, check out the full [documentation](https://huggingface.co/docs/transformers.js).\n\n\n## Installation\n\n\nTo install via [NPM](https://www.npmjs.com/package/@huggingface/transformers), run:\n```bash\nnpm i @huggingface/transformers\n```\n\nAlternatively, you can use it in vanilla JS, without any bundler, by using a CDN or static hosting. For example, using [ES Modules](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Modules), you can import the library with:\n```html\n\n```\n\n\n## Quick tour\n\n\nIt's super simple to translate from existing code! Just like the python library, we support the `pipeline` API. Pipelines group together a pretrained model with preprocessing of inputs and postprocessing of outputs, making it the easiest way to run models with the library.\n\n| Python (original) | \nJavascript (ours) | \n
|---|---|
| \n\n```python\nfrom transformers import pipeline\n\n# Allocate a pipeline for sentiment-analysis\npipe = pipeline('sentiment-analysis')\n\nout = pipe('I love transformers!')\n# [{'label': 'POSITIVE', 'score': 0.999806941}]\n```\n\n | \n\n\n```javascript\nimport { pipeline } from '@huggingface/transformers';\n\n// Allocate a pipeline for sentiment-analysis\nconst pipe = await pipeline('sentiment-analysis');\n\nconst out = await pipe('I love transformers!');\n// [{'label': 'POSITIVE', 'score': 0.999817686}]\n```\n\n | \n