---
license: mit
language:
- pt
library_name: sentence-transformers
pipeline_tag: sentence-similarity
base_model: neuralmind/bert-base-portuguese-cased
tags:
- sentence-transformers
- embeddings
- portuguese
- bertimbau
- mteb
---

# BrazEmbed-PT-BR

A **contamination-clean ~110M Brazilian-Portuguese** text-embedding model, built on the Brazilian
[BERTimbau](https://huggingface.co/neuralmind/bert-base-portuguese-cased) base. This repo is the **single drop-in
model** (TIES-merge of the four task experts): MTEB(por) `mean_16` = **0.6372** (≈ the prior ~100M leader). For the
**headline 0.6567** (the task-routed system, **#1 in the ~100M class**), see 🛠️ [https://github.com/tardellirs/brazembed-pt-br](https://github.com/tardellirs/brazembed-pt-br).

```python
from sentence_transformers import SentenceTransformer
m = SentenceTransformer("tardellirs/brazembed-pt-br")   # mean-pooling, L2-normalized, no instruction prefix
emb = m.encode(["O Brasil é um país tropical da América do Sul."], normalize_embeddings=True)
```

## MTEB(por) — #1 in the ~100M class (routed system 0.6567; this single model 0.6372)

| Model | params | mean_16 |
|---|---:|:---:|
| **BrazEmbed-PT-BR (routed)** | **110M** | **0.6567** |
| BrazEmbed-PT-BR (single, this) | 110M | 0.6372 |
| serafim-100m | 100M | 0.6313 |
| multilingual-e5-small | 118M | 0.6142 |
| granite-embedding-107m | 107M | 0.6080 |

Beats far larger models too (granite-311m 0.6464, multilingual-e5-base 278M 0.6376, serafim-900m 0.6301, …).

## Contamination integrity
All training data was **measured** (hash overlap) to be disjoint from the MTEB(por) test sets (e.g. mMARCO-pt:
0 / 8.8M overlaps vs the Quati pool). A self-audit removed 10 ASSIN2-test pairs that had leaked into our own STS set.
Base BERTimbau is Brazilian (brWaC); training data is Portuguese (a mix of PT-BR-native and PT-translated sources).

## Notes
Development was iterative against MTEB(por) (no held-out split); we use only principled uniform weight-soups, a
no-tuned-knob headline, and report every variant — see [https://github.com/tardellirs/brazembed-pt-br](https://github.com/tardellirs/brazembed-pt-br). License MIT (base BERTimbau). Benchmark:
MTEB(por) ([leaderboard](https://huggingface.co/spaces/MTEB-BR/leaderboard)).