-
Can you please add support for the following quantizations that ctranslate2 added:
int8_float32
int8_bfloat16
float32
bfloat16
-
https://zenn.dev/rinna/articles/5fd4f3cc12f7c5
-
Hello,
I'm trying to convert a model with CTranslate2 4.3 that has been trained with OpenNMT-py 3.5.1 but I get this error:
Converting Saved_Data/Models/fr_en_step_195000.pt to ctranslate2 forma…
-
I have seen some other talk of memory leaks (#390), but I'm having a more sporadic, shorter term issue.
I've experienced this on both an RTX 4070 with 12GB VRAM and an RTX 3090 with 24GB VRAM.
`…
-
**unsupported version**:
1. faster-whisper >= v1.0.0
2. ctranslate2==4.0.0
[From ctranslate2 official guideline that ctranslate2==4.0.0 will support cuda 11.x](https://opennmt.net/CTranslate2/ins…
-
I used
`from faster_whisper import supported_languages`
to import `supported_languages()` function ,I got Error:
```
>>> from faster_whisper import supported_languages
Traceback (most recent ca…
-
### Describe the feature you'd like to request
For visibility. See: https://pypi.org/project/whisper-ctranslate2/
Whisper command line client compatible with original [OpenAI client](https://githu…
-
When I tried to convert `llama-3.2-11b-vision-instruct` downloaded from hugging face using this command:
```bash
ct2-transformers-converter --model /root/llama-3.2-11b-vision-instruct/ --output_dir …
-
I like the idea but it needs more work on execution to be useful IRL.
I would like to see:
- [ ] The ability to choose the model (using arguments)
- [ ] Switch transcription engine to [faster-w…
-
Its a really good model for its size and it aligns with the goal of this project.