Hugging Face

🤗 Hugging Face is the AI community building the future. Build, train and deploy state of the art models powered by the reference open source in machine learning.

Transformers.js

State-of-the-art Machine Learning for the web. Run Hugging Face Transformers directly in your browser, with no need for a server!

The following guide is based on a fresh Minze CLI installation.

1. Install dependencies.

npm

$ npm add @xenova/transformers

yarn

$ yarn add @xenova/transformers

pnpm

$ pnpm add @xenova/transformers

bun

$ bun add @xenova/transformers

2. Create a hf-element.js file inside the src/lib directory.

src/
└─ lib/
   ├─ ...
   └─ hf-element.js

3. Import @xenova/transformers and define a component inside the new file.

src/lib/hf-element.js

import { MinzeElement } from 'minze'
import { pipeline } from '@xenova/transformers'

export class HfElement extends MinzeElement {
  reactive = [['text', 'thinking ...']]

  attrs = ['inputs']

  html = () => `${this.text}`

  async onReactive() {
    const pipe = await pipeline('sentiment-analysis')
    const out = await pipe(this.inputs)

    this.text = JSON.stringify(out[0])
  }
}

4. Add the new element to src/preview.html file.

src/preview.html

<hf-element inputs="I love transformers!"></hf-element>
<my-element>
  <h1>Minze + Vite</h1>
</my-element>

TIP

For more details about Transformers.js refer to the Transformers.js docs.

Huggingface.js

A collection of JS libraries to interact with the Hugging Face API.

The following guide is based on a fresh Minze CLI installation.

@huggingface/inference

A wrapper for the Hugging Face Inference API.

1. Install dependencies.

npm

$ npm add @huggingface/inference

yarn

$ yarn add @huggingface/inference

pnpm

$ pnpm add @huggingface/inference

bun

$ bun add @huggingface/inference

2. Create a hf-element.js file inside the src/lib directory.

src/
└─ lib/
   ├─ ...
   └─ hf-element.js

3. Import @huggingface/inference and define a component inside the new file.

DANGER

Never expose your access token to the public, it should be kept private! If you need to protect it in front-end applications, we suggest setting up a proxy server that stores the access token.

src/lib/hf-element.js

import { MinzeElement } from 'minze'
import { HfInference } from '@huggingface/inference'

const hf = new HfInference('YOUR_HF_ACCESS_TOKEN')

export class HfElement extends MinzeElement {
  reactive = [['text', 'thinking ...']]

  attrs = ['inputs']

  html = () => `${this.text}`

  async onReactive() {
    const result = await hf.textGeneration({
      model: 'gpt2',
      inputs: this.inputs
    })

    this.text = result.generated_text
  }
}

4. Add the new element to src/preview.html file.

src/preview.html

<hf-element inputs="The answer to the universe is"></hf-element>
<my-element>
  <h1>Minze + Vite</h1>
</my-element>

TIP

For more details about Huggingface.js refer to the Huggingface.js docs.