vexira

Vexira

An 80.8M-parameter TR↔EN translation model. Runs fully offline on CPU — no internet, no API key, no GPU. Built for subtitles, game text, UI strings and file content.

Code and full documentation: https://github.com/uixova/vexira

Benchmarks

FLORES-200 devtest (1012 sentences, beam 4, glossary off):

For reference on the same set: NLLB-200 distilled (600M) ≈ 26–28 en→tr, opus-mt-tc-big-en-tr (230M) ≈ 30. Vexira reaches that band with 80.8M parameters.

Term consistency — measured on a real 559-line Ren'Py game file, counting terms rendered differently across lines (a defect BLEU cannot see):

The built-in glossary additionally resolves 42% of those strings by exact match, where the model is never invoked at all.

Files

Both checkpoints carry the tokenizer and a 1,234-term glossary embedded inside the weights file, so a single .pt is self-contained.

Usage

This is a standalone PyTorch model, not a transformers architecture. Use the repository code:

git clone https://github.com/uixova/vexira && cd vexira
pip install torch sentencepiece
huggingface-cli download uixova/vexira vexira_sft.pt vexira_spm.model --local-dir models/

python translate.py --text "Hello world" --to tr
# -> Merhaba dünya

from translate import Translator

tr = Translator()
print(tr.translate(["Save", "Are you sure you want to quit?"],
                   tgt_lang="tr", domain="ui"))
# ['Kaydet', 'Çıkmak istediğinizden emin misiniz?']

domain selects the register: sub subtitles · ui interface · doc prose · ocr screen text. With ui the glossary makes terminology deterministic.

Architecture

Encoder-decoder transformer, 12 encoder + 6 decoder layers, d_model 512, d_ff 1408, 8 heads, RMSNorm (pre-LN), SwiGLU, learned absolute positions, tied embeddings, 32k shared SentencePiece unigram vocabulary with byte_fallback.

No quantization: weights are fp16, computed in fp32 on CPU.

Limitations

• 128-token effective context. max_pos is 512 but positions 128–511 were never trained (pretraining used max_len=128); the code measures this and splits longer input at sentence boundaries, chaining parts with a context token. Measured: 16 sentences / 144 tokens preserved 16/16 concepts.
• TR↔EN only.
• Not a knowledge or chat model. It translates; it does not answer questions or reason.
• Poetry, wordplay, heavy idiom and slang are weak.
• The reserved <ocr>, <stt>, <tts> domain tokens exist but are untrained.

Training

Pretraining: 60.9M sentence pairs → 121.7M bidirectional examples, 7.56B tokens total (4.67× Chinchilla for this size), 4 sessions on 2×T4.

Fine-tuning: 133,753 examples, 3 epochs at LR 5e-5. The set is deliberately balanced — 51% UI / 49% general, with a length distribution matching pretraining — so that UI terminology is learned without losing general translation ability.

Data: 23 OPUS corpora (OpenSubtitles, CCMatrix, HPLT, CCAligned, WikiMatrix, TED, Tatoeba, KDE4, SETIMES and others) plus system locale catalogs for UI terminology. FLORES-200 devtest was never part of training.

Citation

@misc{vexira2026,
  title  = {Vexira: an 80M-parameter Turkish-English translation model},
  author = {uixova},
  year   = {2026},
  url    = {https://github.com/uixova/vexira}
}