Zero Shot Nlp Classifier Multi Lang

Zero Shot Nlp Classifier Multi Lang

Sorts text into labels you choose, without training, across languages, using several named zero-shot models in a Gradio page. For analysts, researchers and NLP developers.

Made by iamkb on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This Gradio app classifies text into categories you supply, with no training step. That approach is called zero-shot classification. The record names six models, which by their names include multilingual language models and a language identification model.

Who it suits: It suits analysts, researchers and developers who need to label text in more than one language and want to compare several models on the same input. The record gives no notes on the controls, so expect to learn the page by clicking.

Industry | Who uses it | How they use it | Example
Research | Text analysts | Label texts with categories they define | Enter a sentence and a few candidate labels
Customer support | Support teams | Sort messages by topic without training a model | Try labels such as billing or delivery on a message
Software | NLP developers | Compare several zero-shot models on one input | Run the same text through different models

You can skip it if you have labeled data and want a trained classifier.

Where to use it:

• A web browser, on the Space's page on Hugging Face
• Windows, macOS or Linux with Python, from a downloaded copy

Use case | Who | What they get back
Label text with your categories | Analysts | Scores for each label you gave
Test multilingual input | Researchers | A view of how models handle other languages
Compare models | Developers | Outputs from several named models

How to set it up:

• Nothing at all if you use the online Space
• Python on your computer, only if you run a copy
• The packages listed in requirements.txt

1. Open the Space's page in a browser to use it with nothing to install.
2. To run a copy, install the Hugging Face Hub package and download the Space.
3. Move into the new folder, then install Gradio and the listed requirements.
4. Start the app with Python.

pip install -U huggingface_hub
hf download iamkb/zero-shot-nlp-classifier-multi-lang --repo-type space --local-dir zero-shot-nlp-classifier-multi-lang
cd zero-shot-nlp-classifier-multi-lang
pip install gradio
pip install -r requirements.txt
python app.py

How to use it: Online, open the Space and use the controls on its page. On your own machine, run the main file from the Space folder, then use the page it serves. The record does not list the controls, so look at the page.

python app.py

Example: You open the Space, enter a sentence and a few labels you made up. The app returns how well each label fits the sentence. The record shows no sample, so the exact inputs depend on the page.

For newcomers and experts: Normally a classifier must be trained on examples of each label. Zero-shot classification skips that, so you can type your own labels and get a ranking at once. Start with a short sentence and two or three clear labels.

If you know the field, most of the named models are natural language inference models, which score whether a label follows from the text. One is a language identifier. Results depend on the model and the wording of your labels, and the record gives no accuracy figures.

You enter text and candidate labels
The page may detect the language
A chosen model scores each label
The page shows the ranking

Good to know:

• License: MIT, as listed on the Space
• Cost to run: none named
• Data it reaches: the text and labels you enter, and the six models it names
• Not for: the record names no exclusions, and it gives no accuracy figures

On Hugging Face, 13 November 2026

• Likes: 5
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/iamkb/zero-shot-nlp-classifier-multi-lang
• Files: https://huggingface.co/spaces/iamkb/zero-shot-nlp-classifier-multi-lang/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.