← Directory

libretranslate

Free and Open Source Machine Translation API, entirely self-hosted.

  • AI
  • Other
  • Productivity
  • Tools & Utilities
Free and Open Source Machine Translation API, entirely self-hosted. You will need to create a "cache" and a "share" folder manually in the specified appdata folder manually before the container will run. First start takes a while as models are being downloaded.

Platform availability

Available on 5 of 9 platforms

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

32/100 Low

Maintained
0GitHub repository no longer exists
Popular
No data
Easy to install
755 install platforms, install notes, Docker image
Light to run
602 GB minimum RAM, runs on ARM

Based on 3 of 4 factors.

Availability: platform listings confirmed today, links checked 15 days ago

No GitHub repository data collected yet.

Checked 14 days ago - source: GitHub (BigBearTechWorld/libretranslate)

Resources & Compatibility

Minimum requirements: 2 GB RAM. ARM (e.g. Raspberry Pi) is supported.

Checked 14 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Set valid storage before starting Point the app at a valid storage address such as memory:// or a Redis location (redis://host:6379/0), because built-in background tasks need it to start. If the address is invalid, startup fails and features that rotate secrets or block abusive requests won't work. source
  • Start without debug for real use Run it without the debug flag for normal use so it uses the production server (waitress) instead of the developer test server. Debug mode is only for testing and can be less safe or stable. source
  • Load only the languages you use Limit startup to the language models you actually need, and avoid automatic model downloads unless you want them. This saves memory and stops large surprise downloads on first run. source

Fetched 14 days ago - each note links to its source

Alternatives

No alternatives collected yet.

Common questions

What storage setting is required before LibreTranslate starts?

Point the app at a valid storage address such as memory or a Redis location, because built-in background tasks need it to start. If the address is invalid, startup fails and features that rotate secrets or block abusive requests won't work.

Should I run LibreTranslate with debug enabled in production?

No, run it without the debug flag for normal use so it uses the production server instead of the developer test server. Debug mode is only for testing and can be less safe or stable.

How can I avoid large model downloads and high memory use at startup?

Load only the language models you actually need, and avoid automatic model downloads unless you want them. This saves memory and prevents large surprise downloads on first run.

What common Docker issues can cause LibreTranslate to stop working after startup?

Common causes include memory limits, model downloads, GPU binding, port conflicts, and slow inference. These are the actual errors the article fixes with working remedies.

Answers sourced from github.com, github.com, github.com, mustafa.net

Community

Collected 8 days ago - public community threads