Instructions to use WindstormLabs/translate-pt-en with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use WindstormLabs/translate-pt-en with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("translation", model="WindstormLabs/translate-pt-en")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("WindstormLabs/translate-pt-en", dtype="auto") - Notebooks
- Google Colab
- Kaggle
File size: 2,146 Bytes
2c21ef0 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 | ---
license: cc-by-4.0
tags:
- translation
- marian
- windyword
- portuguese
- english
language:
- pt
- en
library_name: transformers
pipeline_tag: translation
---
# WindyWord.ai Translation — Portuguese → English
**Translates Portuguese (`pt`) text into English (`en`).**
## Quality & Model Lineage
- **Tier:** Standard ⭐⭐⭐⭐
- **Base model:** [Helsinki-NLP/opus-mt-ROMANCE-en](https://huggingface.co/Helsinki-NLP/opus-mt-ROMANCE-en) — 68k+ downloads, multilingual Romance-source model
- **Architecture:** MarianMT
- **Note:** Helsinki-NLP does not publish a dedicated Portuguese-only `opus-mt-pt-en` model. Their canonical path for Portuguese-source translation is the multilingual `opus-mt-ROMANCE-en`, which handles Portuguese alongside Italian, Spanish, French, Catalan, and Romanian as source languages. This WindyWord repo wraps the same underlying weights with a Portuguese-specific repo name to satisfy installer / consumer code that expects a `translate-pt-en` artifact.
For sibling Romance-source pairs use [WindyWord/translate-ROMANCE-en](https://huggingface.co/WindyWord/translate-ROMANCE-en) directly (same underlying model).
## Available Variants
| Variant | Description |
|---|---|
| `lora/` | **WindyStandard** — production baseline. Helsinki-NLP/opus-mt-ROMANCE-en weights wrapped under the WindyWord pt-en namespace. |
## Quick Usage
```python
from transformers import MarianMTModel, MarianTokenizer
tokenizer = MarianTokenizer.from_pretrained("WindyWord/translate-pt-en", subfolder="lora")
model = MarianMTModel.from_pretrained("WindyWord/translate-pt-en", subfolder="lora")
```
## Commercial Use
Visit [windyword.ai](https://windyword.ai) for apps and API access.
---
## Provenance & License
Weights derived from [Helsinki-NLP/opus-mt-ROMANCE-en](https://huggingface.co/Helsinki-NLP/opus-mt-ROMANCE-en) under CC-BY-4.0. Wrapped as a Portuguese-specific entry by WindyWord 2026-04-28 to support installer expectations for a `translate-pt-en` repo. Underlying model is identical to `WindyWord/translate-ROMANCE-en`.
*Certified by Opus 4.6 Opus-Claw (Dr. C) on Veron-1 (RTX 5090, Mt Pleasant SC).*
|