Translation
Transformers
Safetensors
French
Yoruba
marian
text2text-generation
candle
opus-mt
malinali
# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("malinali-app/opus-mt-fr-yo")
model = AutoModelForSeq2SeqLM.from_pretrained("malinali-app/opus-mt-fr-yo", device_map="auto")Quick Links
fr → yo Marian / Candle (Malinali)
Public on-device pack for Malinali: Helsinki-NLP OPUS-MT
weights in safetensors form plus fast tokenizers for Candle (marian_flutter).
Upstream credit
- Base model:
Helsinki-NLP/opus-mt-fr-yo - Project: Helsinki-NLP / OPUS-MT
- License: follow the upstream model card (typically CC-BY 4.0 for OPUS-MT)
Malinali only repackages weights + converts SentencePiece → Hugging Face fast tokenizer JSON for on-device inference. We do not claim ownership of the trained model.
Files (required)
| File | Role |
|---|---|
config.json |
Marian config |
model.safetensors |
Weights |
tokenizer-enc.json |
Source fast tokenizer |
tokenizer-dec.json |
Target fast tokenizer |
Direction: fr → yo.
- Downloads last month
- 27
Model tree for malinali-app/opus-mt-fr-yo
Base model
Helsinki-NLP/opus-mt-fr-yo
# 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="malinali-app/opus-mt-fr-yo")