search

OpenNMT / CTranslate2

Fast inference engine for Transformer models
https://opennmt.net/CTranslate2
MIT License
3.43k stars 306 forks source link

Python process crashes on exit under Windows with CUDA #1782

Open TechInterMezzo opened 2 months ago

TechInterMezzo commented 2 months ago

I am using Windows 10 with CUDA 11.8, Python 3.10 and ctranslate2 4.4.0. The problem already seems to reported in the faster-whisper repo but I also have it when I directly use ctranslate2: https://github.com/SYSTRAN/faster-whisper/issues/71

The python process crashes after executing this code:

import ctranslate2
model = ctranslate2.models.Whisper("openai/whisper-tiny", device="cuda")

It is not the the code itself that crashes the process but I guess some kind of internal routine on process exit that is necessary for CUDA.

minhthuc2502 commented 2 months ago

Try with CUDA 12.x . Ctranslate2 is compatible with cuda 12

nebehr commented 3 weeks ago

I do not use ctranslate2 directly, but I do use faster-whisper with CUDA 12.5, and the issue is reproducible almost always unless it's with very short audio/video files. It does not necessarily mean that file size is the culprit, it's just that with smaller inputs it sometimes does complete without a crash, but with any decent-sized one it always crashes.