locAGI is an integrated cloud CAT platform for enterprise localisation teams. The workbench itself is free — sign up, invite your linguists, and start translating. You pay only for the AI tokens you consume and the storage you actually use. Or skip our billing entirely and bring your own AI keys (OpenAI / Anthropic / Google) — your call.
Two options — pick what works for your team:
Both paths give linguists the same integrated editor experience. Switch between them any time.
Traditional CAT tools charge per-seat subscriptions or thousand-euro perpetual licenses, then put AI behind a separate plugin or hosted-MT marketplace. locAGI flips both halves of that.
Yes. No credit card to create a workspace, invite your linguists, build termbases, or use the integrated CAT editor. You only enter payment details if you choose to buy AI tokens through us or exceed the included storage. You can also stay 100% free by bringing your own AI keys and keeping your project sizes within the free tier.
You connect your own OpenAI / Anthropic / Google key in Settings. locAGI sends the API call straight to that vendor; the card you have on file with them is the one charged. We don’t see your AI bill or take a cut. BYOK is for teams that already have provider relationships, procurement contracts, or in-house spending controls.
One bill, no procurement work, no managing multiple vendor accounts. Useful for smaller teams that don’t want to negotiate enterprise rates with three different AI providers. You can mix — some users on BYOK, others on pay-as-you-go — in the same workspace.
No. locAGI is a managed cloud service. We considered self-hosting but decided against it — security reviews, version drift, support overhead, and the AI marketplace work all benefit from a single managed deployment. If air-gapped or fully on-prem is a hard requirement for your programme, let us know and we’ll tell you honestly whether we’re a fit yet.
No. locAGI is a proprietary commercial product. We invest heavily in the integrated CAT, AI alignment, and security posture; keeping the code closed is what funds that work. Your data remains yours and is fully exportable (XLIFF, TMX, TBX, MQXLIFF) at any time — no lock-in on your translations, even if the source is closed.
You swap to a cheaper one in Settings — no migration, no lost work. The per-model cost projection in the Studio analysis modal lets you see the impact of the change before you commit. Both BYOK and pay-as-you-go support every provider we’ve integrated.