Downloads · 30 days
6
18% of all-time downloads
translate-windy-hplt-bg-en
WindyWord/translate-windy-hplt-bg-en
translate-windy-hplt-bg-en is a translation model from WindyWord. Use it when you need text moved from one language to another. It is set up for ctranslate2. The card lists the license as cc-by-4.0.
Part of the Windstorm Labs open model catalogue. Scores, licence and attribution for every model: https://windytranslate.com/models/translate-windy-hplt-bg-en Canonical copy: https://huggingface.co/WindyTranslate/tran…
All-time downloads
33
Public
Repo size
80.7 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin79 MB · 90%
At a glance
- Task
- Translation
- Library
- ctranslate2
- License
- cc-by-4.0
- Access
- Public
- Created
- Aug 4, 2026
- Updated
- Oct 1, 2026
- SHA
- 07362235
Try a prompt
Base models
Model card
From the Hugging Face model README
translate-windy-hplt-bg-en — bg to en
Part of the Windstorm Labs open model catalogue. Scores, licence and attribution for every model: https://windytranslate.com/models/translate-windy-hplt-bg-en
Canonical copy: https://huggingface.co/WindyTranslate/translate-windy-hplt-bg-en
CTranslate2 INT8 bilingual translation model, derived from
HPLT/translate-bg-en-v2.0-hplt_opus (the HPLT project, CC-BY-4.0).
CC-BY-4.0 permits commercial use and modification and requires attribution; this notice provides it. Windstorm Labs did not train the base model.
Recovery note
HPLT publishes native Marian (.npz + .spm + .vocab) and marks its repos
inference: false. An earlier conversion of this model to transformers format corrupted the
weights, producing untranslatable output, and the model was written off. It was recovered on
2026-07-26 by converting the original .npz with the official Marian converter. The model
was always fine; the conversion was not.
What was changed
A deliberately minimal LoRA fine-tune ("windy hallmark") merged into the base weights.
| Rank / alpha | 8 / 16 |
| Learning rate / steps | 2.5e-06 / 50 |
| Target modules | q_proj, v_proj |
| Seed | 42 (reproducible) |
| Training data | opus100_bg-en, 2000 sentence pairs |
| Tensors modified | 36 |
| Max absolute weight delta | 4.183e-05 |
Provenance
base model.bin sha256 d3132e6ebe69b21e27103589c414c85ea12b50c2e758812adae1feb7884d166e
this model.bin sha256 9e15e2e9d07d1140e06c1be0263618e46c2e5bd06c9f8cb8bd51146478a07afe
Verified on the quantized artifact you download, not merely internal weights.
Evaluation
FLORES-200 devtest, 200 sentences, beam 4, CTranslate2 INT8 on CPU.
Metrics: spBLEU (flores200 tokenizer) + chrF (word_order=0).
| spBLEU | chrF | |
|---|---|---|
| this model | 35.96 | 61.87 |
| upstream, pre-fog | 35.69 | — |
Limitations
- One language pair only (bg to en).
- FLORES-200 is news and encyclopedic prose; it says little about conversational register.
- No human evaluation was performed.
- Task
- Translation
- Library
- ctranslate2
- License
- cc-by-4.0
- Created
- Aug 4, 2026
- Updated
- Oct 1, 2026