kevinveenbirkenbachandClaude Opus 5 1832bbfe6f feat(i18n): translate on the GPU when the host can hand one to a container ...

The ephemeral LibreTranslate container starts against the -cuda image with
--gpus all wherever docker registers an nvidia runtime, and keeps the CPU
image everywhere else, so runners without the container toolkit behave as
before. Measured on an RTX 3080 Ti: 44% utilisation, 1744 MiB resident,
482 sentences per 30 seconds.

Only the tool derives the suffix. web-svc-libretranslate stays pinned to the
CPU image, so no deploy host is required to install the toolkit.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

2026-09-23 02:26:23 +02:00

..

cli

feat(i18n): gettext catalogs with LibreTranslate machine translation

2026-09-22 18:22:03 +02:00

group_vars

fix(litellm): publish a backend only when the gateway's own flag enables it

2026-09-15 02:06:38 +02:00

library

…

plugins

Merge branch 'main' into feature/store-i18n

2026-09-22 21:37:33 +02:00

roles

Merge branch 'main' into feature/store-i18n

2026-09-22 21:37:33 +02:00

scripts

fix(diagnostics): let the in-zone probes read the drop-in that holds the zone

2026-09-03 12:50:58 +02:00

utils

feat(i18n): translate on the GPU when the host can hand one to a container

2026-09-23 02:26:23 +02:00

__init__.py

…