Connectors
Push-model companions that translate content in place in external systems, such as Airtable, Jira, Strapi, Contentful, Shopify and WordPress, and write each language back as the source system's own version.
Docs · Features
Every feature, one page each.
Localization OS is built from a set of focused capabilities that share one pipeline. Each page below covers what it does, how to get started, what it works with, and its current limits, stated plainly.
Data export
A checksummed, access-confined export of a scope's whole object graph, one JSONL file per entity plus a manifest, in a ZIP, run as an async job you start, poll and download.
Documents and the editor
Upload a file, translate it, review the segments a person needs to look at, and export it back in its original format.
Engines and providers
The models that do the work, your local server, a cloud LLM, or a machine-translation service, and every step, judge and extraction resolves one.
Evaluations Lab
A sandbox for one question, which setup translates better, answered with statistics that refuse to declare a winner when the evidence is thin.
Glossaries and terminology
The terminology authority: the list of terms that must be translated a particular way, or must not be translated at all, enforced rather than merely suggested.
Image localization (Beta)
Upload an image and get it back with its text in another language. Detect the text regions, translate them, and re-render the image, on the ordinary document pipeline.
Key-based projects
A second kind of project for continuous localization: the unit of work is a string key, and the deliverable is one key file per language, on demand, with a change feed.
AI agents and MCP
How an AI agent, script, or automated integration drives Localization OS through the API, the CLI, and a purpose-built MCP server.
Memory
The system's notebook about one customer: terminology decisions and style rules it keeps applying, some written by you, some learned from corrections people make.
Queries
Threaded questions and comments on a project, document, segment or string key, visible to everyone who can see the project, with an open-to-closed lifecycle.
Reports
Operational cost, token, volume, QA and post-edit reporting over the data you can already see: a dashboard plus a report builder that exports to CSV.
Content routing
Inbound routing rules that match an incoming submission and choose where it lands and which engine translates it.
Translation memory
Every approved translation is remembered, so the same sentence never has to be translated, or paid for, twice.
Vendors
First-class external translation and review vendors: a workflow can dispatch its packaged text to a human vendor and wait for a signed return.
Workflows
A recipe: an ordered list of steps a document runs through, from translation to automated quality checks to human review.
Worklist
The cross-project reviewer worklist: one page listing the paused review steps, queries and documents that need your attention.