Phase 1 initial upload: Helsinki-NLP/opus-mt-cpp-en → CT2 float16

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+source.spm filter=lfs diff=lfs merge=lfs -text
+target.spm filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,36 @@
+---
+license: apache-2.0
+tags:
+- translation
+- opus-mt
+- ctranslate2
+- custom
+source_model: Helsinki-NLP/opus-mt-cpp-en
+---
+
+# ooeoeo/opus-mt-cpp-en-ct2-float16
+
+CTranslate2 float16 quantized version of `Helsinki-NLP/opus-mt-cpp-en`.
+
+Converted for use in the [ooeoeo](https://ooeoeo.com) desktop engine
+with the `opus-mt-server` inference runtime.
+
+## Source
+
+- Upstream model: [Helsinki-NLP/opus-mt-cpp-en](https://huggingface.co/Helsinki-NLP/opus-mt-cpp-en)
+- Variant: `custom`
+- Language pair: `custom`
+
+## Conversion
+
+```bash
+ct2-transformers-converter \
+    --model Helsinki-NLP/opus-mt-cpp-en \
+    --output_dir ct2/ \
+    --quantization float16 \
+    --copy_files source.spm target.spm tokenizer_config.json vocab.json
+```
+
+## License
+
+Apache-2.0 (inherited from upstream Helsinki-NLP/Opus-MT).

config.json

@@ -0,0 +1,10 @@
+{
+  "add_source_bos": false,
+  "add_source_eos": false,
+  "bos_token": "<s>",
+  "decoder_start_token": "</s>",
+  "eos_token": "</s>",
+  "layer_norm_epsilon": null,
+  "multi_query_attention": false,
+  "unk_token": "<unk>"
+}

model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3217b35cd94cf81dd6b1d512e32187bfa7852608869d1a55224a645810eabb50
+size 148104129

source.spm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:921eac957d02068ba26ce8f413c8d67a8912d8d619da5901e011ef1afe5aaa5d
+size 814512

target.spm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fa9f577c9b97fe30fb2257e3873faa95a9caeca596faee22811d40e7afae0ea
+size 811797

tokenizer_config.json

@@ -0,0 +1 @@
+{"target_lang": "eng", "source_lang": "cpp"}