Open-weight model
marian-mt-bbc-en-pcm
by National Information Technology Development Agency Hub AI Research Team NITHUB-AI/marian-mt-bbc-en-pcm
marian-mt-bbc-en-pcm is an open-weight model from National Information Technology Development Agency Hub AI Research Team, released under Creative Commons Attribution 4.0. It has 512-token context. Its published files total 298.8 MB. It draws 229 downloads a month.
Parallel Sentences from the message translation (English) and Pidgin translation of the Bible.
Model Card
By National Information Technology Development Agency Hub AI Research Team, published under cc-by-4.0, revision 84025d80f625.
Parallel Sentences from the message translation (English) and Pidgin translation of the Bible. - Pidgin sentences from BBC Pidgin and English translations by GPT3.5-turbo TBA Bridging the Gap with Retrieval-Augmented Generation: Making Prosthetic Device User Manuals Available in Marginalised Languages
Read National Information Technology Development Agency Hub AI Research Team's full model card
- source language: en (English)
-
target language: pcm (Nigerian Pidgin)
-
dataset:
- Parallel Sentences from the message translation (English) and Pidgin translation of the Bible.
- Pidgin sentences from BBC Pidgin and English translations by GPT3.5-turbo
- model: transformer-align
- pre-processing: normalization + SentencePiece
Performance
TBA
Creator
- Fortune Adekogbe
Model in Action
- Bridging the Gap with Retrieval-Augmented Generation: Making Prosthetic Device User Manuals Available in Marginalised Languages
- Code: https://github.com/Iykay/User-Manual-LLM
Configuration
- Architecture
- MarianMTModel
- Context length (tokens)
- 512
- Layers
- 6
- Vocabulary size
- 58,101
- Stored precision
- float32
- Model type
- marian
Identity and Version
- Repository
- NITHUB-AI/marian-mt-bbc-en-pcm
- Publisher
- National Information Technology Development Agency Hub AI Research Team
- Task
- Not stated by the source
- Modality
- Other
- Library
- transformers
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 84025d80f6253c2d27e5409950a0f165f7665824
- First published
- 2023-09-04
- Last updated
- 2026-10-16
Files and Weights
10 files, 298.8 MB in total. The weights are 1 file totalling 295.9 MB in bin.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| pytorch_model.bin | Weights | 295.9 MB | 47b482fc7414 |
| config.json | Configuration | 1.4 KB | — |
| model_args.json | Configuration | 3.0 KB | — |
| special_tokens_map.json | Configuration | 74 B | — |
| README.md | Documentation | 648 B | — |
| source.spm | Other | 768.5 KB | — |
| target.spm | Other | 796.8 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| tokenizer_config.json | Tokenizer | 280 B | — |
| vocab.json | Tokenizer | 1.4 MB | — |
License and Download
- License
- cc-by-4.0
- Access
- Open weights, no gate
- Download size
- 295.9 MB
Released by National Information Technology Development Agency Hub AI Research Team through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 295.9 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About marian-mt-bbc-en-pcm
Can I use marian-mt-bbc-en-pcm commercially?
Yes. marian-mt-bbc-en-pcm is released under Creative Commons Attribution 4.0. CC BY 4.0 permits sharing and adapting the work, including commercially, provided the creator is credited and changes are indicated.
What is marian-mt-bbc-en-pcm's context length?
512 tokens, from the maximum position embeddings in its published configuration.