README.md

---
library_name: transformers
language:
- en
- fr
- it
- es
- ru
- uk
- tt
- ar
- hi
- ja
- zh
- he
- am
- de
license: openrail++
datasets:
- textdetox/multilingual_toxicity_dataset
metrics:
- f1
base_model:
- cis-lmu/glot500-base
pipeline_tag: text-classification
tags:
- toxic
---

## Multilingual Toxicity Classifier for 15 Languages (2025)

This is an instance of [Glot500](https://huggingface.co/cis-lmu/glot500-base) that was fine-tuned on binary toxicity classification task based on our updated (2025) dataset [textdetox/multilingual_toxicity_dataset](https://huggingface.co/datasets/textdetox/multilingual_toxicity_dataset).

Now, the models covers 15 languages from various language families:

| Language  | Code | F1 Score |
|-----------|------|---------|
| English   | en   | 0.9071  |
| Russian   | ru   | 0.9022  |
| Ukrainian | uk   | 0.9075  |
| German    | de   | 0.6528  |
| Spanish   | es   | 0.7430  |
| Arabic    | ar   | 0.6207  |
| Amharic   | am   | 0.6676  |
| Hindi     | hi   | 0.7171  |
| Chinese   | zh   | 0.6483  |
| Italian   | it   | 0.5975  |
| French    | fr   | 0.9125  |
| Hinglish  | hin  | 0.7051  |
| Hebrew    | he   | 0.8911  |
| Japanese  | ja   | 0.9058  |
| Tatar     | tt   | 0.5834  |

## How to use

```python
import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained('textdetox/glot500-toxicity-classifier')
model = AutoModelForSequenceClassification.from_pretrained('textdetox/glot500-toxicity-classifier')

batch = tokenizer.encode("You are amazing!", return_tensors="pt")

output = model(batch)
# idx 0 for neutral, idx 1 for toxic
```

## Citation
The model is prepared for [TextDetox 2025 Shared Task](https://pan.webis.de/clef25/pan25-web/text-detoxification.html) evaluation.

Citation TBD soon.