---
language:
  - multilingual
tags:
  - model2vec
  - embeddings
  - european
license: mit
base_model: intfloat/multilingual-e5-small
---

# m2v-e5-small-european

A [Model2Vec](https://github.com/MinishLab/model2vec) static embedding model distilled from [intfloat/multilingual-e5-small](https://huggingface.co/intfloat/multilingual-e5-small) (118M params), pruned to European languages only.

Pruned 36.5% of tokens (removed CJK, Arabic, Hebrew, Thai, Devanagari, Korean, Japanese, etc.).

| | Before pruning | After pruning |
|---|---|---|
| Vocabulary | 249,999 tokens | 158,843 tokens |
| Embedding dim | 256 | 256 |

## Usage

```python
from model2vec import StaticModel
model = StaticModel.from_pretrained("flipbitsnotburgers/m2v-e5-small-european")
embeddings = model.encode(["deodorant", "Duschgel", "shower gel"])
```

## License

MIT (same as base model)