libctranslate2-4 (4.8.2-1~cuda13)

Published 2026-09-25 14:03:26 +00:00 by eric

Installation

distribution=
component=
sudo curl https://git.lemesre.org/api/packages/eric/debian/repository.key -o /etc/apt/keyrings/gitea-eric.asc

echo "deb [signed-by=/etc/apt/keyrings/gitea-eric.asc] https://git.lemesre.org/api/packages/eric/debian $distribution $component" | sudo tee -a /etc/apt/sources.list.d/gitea.list

sudo apt update
sudo apt install libctranslate2-4=4.8.2-1~cuda13

Repository Info

Distributions
jammy, noble
Components
main
Architectures
arm64

About this package

fast inference engine for Transformer models (CUDA 13, arm64) CTranslate2 is a C++ and Python library for efficient inference with Transformer models (Whisper, NLLB, T5, Llama, ...), with weight quantization and CPU/GPU execution. . This build enables the CUDA backend for Linux arm64 (Jetson Orin, Grace Hopper, DGX Spark GB10), which the upstream project does not publish for this architecture. CUDA libraries are loaded at run time (cuBLAS 13, cuDNN 9) from the CUDA toolkit or from the nvidia-* pip packages; without them the library still works on CPU (OpenBLAS + Ruy).

Dependencies

libc6 (>= 2.34)
libgcc-s1 (>= 4.5)
libgomp1 (>= 4.9)
libopenblas0
libstdc++6 (>= 12)
Details
Debian
2026-09-25 14:03:26 +00:00
4
Eric Lemesre
6.6 MiB
Assets (2)
Versions (2) View all
4.8.2-1~cuda13 2026-09-25
4.6.3-1~cuda13 2026-09-24