You can deploy this model and start earning money today!

WindyWord/
translate-windy-hplt-en-tr

Name: translate-windy-hplt-en-tr
Availability: InStock

HPLT/translate-en-tr-v2.0-hplt_opus

finetune

cc-by-4.0

Views

No views yet

translate-windy-hplt-en-tr — en to tr

CTranslate2 INT8 bilingual translation model, derived from HPLT/translate-en-tr-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_en-tr (clinic cache), 2000 sentence pairs
Tensors modified	36
Max absolute weight delta	4.079e-05

Provenance

base   model.bin sha256  b3821ec5bd4db96d3214fea6ac5e5aeccb5328802f54fceea357b91995c95c81
this   model.bin sha256  37d77d3eedc0591b1c01ca2d0838f7901972ae99db42ccf86c843d5bb5ba784b

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	36.93	58.14
upstream, pre-fog	36.61	—

Limitations

One language pair only (en to tr).
FLORES-200 is news and encyclopedic prose; it says little about conversational register.
No human evaluation was performed.

WindyWord/translate-windy-hplt-en-tr