NMT Marian Framework Demo

NMT Marian Framework Demo is a self-hosted demonstration of the Marian neural machine translation (NMT) framework that runs on your own hardware, translating…

  • Tools & Utilities
NMT Marian Framework Demo is a self-hosted demonstration of the Marian neural machine translation (NMT) framework that runs on your own hardware, translating text between languages without external services. It is ideal for organizations needing secure, high-quality translations while keeping their data completely confidential.

Platform availability

Available on 0 of 10 platforms

  • Unraid (not listed)
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)
  • Any server (not listed)

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
0last commit 1369 days ago
Popular
631,476 GitHub stars
Easy to install
250 install platforms, install notes, Docker image
Light to run
No data
Security check
65no serious findings, 1 minor, CI, tests, OpenSSF Scorecard 3/10

Based on 4 of 5 factors.

GitHub stars
1,476
Open issues
97
Last commit
2023-02-21
Latest release
1.12.0
Activity
stale

Checked 16 days ago - source: GitHub (marian-nmt/marian)

Resources & Compatibility

ARM (e.g. Raspberry Pi) is not supported.

Checked 54 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Config files use INI format Marian NMT config files are INI-style, so lines starting with ; are comments and square-bracket sections like [default] group settings. Put each option in the right section, or the app may ignore it and run with defaults. source

Fetched 53 days ago - each note links to its source

Alternatives

No alternatives collected yet.

Common questions

How should Marian config files be formatted?

Marian config files use INI format, where lines starting with ; are comments and square-bracket sections such as [default] group settings. Put each option in the correct section, or the app may ignore it and run with defaults.

How can I set common model architecture and training options?

Marian provides --task options as a shortcut for setting model architecture and training options for common NMT model configurations. The documentation lists the predefined configurations.

How do I compile Marian with a custom feature set?

The Marian build system uses CMake to compile the framework with various configurations. It supports cross-platform builds and fine-grained control over features and optimizations.

What guidance is available for using custom Marian models with Hugging Face Transformers?

A Hugging Face discussion asks how to use custom Marian-NMT models, including models from the Opus-MT repository, with Transformers.

Answers sourced from deepwiki.com, discuss.huggingface.co, github.com, marian-nmt.github.io

Community

Collected 48 days ago - public community threads