AI providers

Your models. Your profiles. Your choice.

Omni builds on the official Umbraco.AI, giving agencies and site owners the freedom to choose supported providers and make approved profiles available to editors.

No single-provider lock-in

Choose the service that fits the project.

Connect a supported Umbraco.AI chat provider—such as OpenAI, Anthropic, Google Gemini or Amazon Bedrock—and Omni uses the standard abstraction rather than a proprietary integration.

OpenAIAnthropicGoogle GeminiAmazon BedrockMicrosoft AI FoundryAlibaba Cloud Model StudioMoonshot AIOpenRouterZ.AI

Choose a chat provider package compatible with your Umbraco version and a model suited to your source and target languages. Umbraco.AI provider packages are free to install, but model API tokens and provider usage charges are separate and are not included with Omni. Learn about provider setup and costs.

Explore Umbraco.AI providers ↗
Configure Umbraco.AI and create translation profiles for Omni.Playing this video loads YouTube and may set cookies. Video privacy details.
Administrators choose which live chat profiles translators can use.
Omni settings for allowed AI chat profiles

Administrators choose which live chat profiles translators can use.

Multiple profiles

Give editors useful choice without losing governance.

Administrators can whitelist AI profiles and set a default. Editors then choose an approved profile for a Dictionary, page or whole-section job.

  • Different models for speed, quality or cost
  • Different prompts for brands, clients or content styles
  • A central default for everyday translation
  • Live validation if a profile is removed or changed

For agencies

Adapt the AI layer without changing the editor workflow.

Standardise one setup across clients, select providers to match their requirements and keep the friendly Omni experience consistent for content teams.

Where your content goes

Choose the provider that handles your text.

Omni sends the text you select to the Umbraco.AI provider configured on your own site, using your own credentials. Nothing is routed through a Diplo service.

  • No translation content is sent to Diplo
  • Translation requests go through your site's Umbraco.AI setup
  • Provider credentials are managed through your own Umbraco.AI setup
  • Omni extracts selected, supported text instead of sending the full page structure

Translation requests also include instructions and matching glossary rules. Your provider's terms govern how it handles requests, including retention and training policies. Your site's Umbraco.AI audit, logging and diagnostics settings may also retain request or response data. Choose these settings to suit your organisation's requirements.

Diplo Translator Omni

Ready to translate your first page?

Bring AI translation into the Umbraco backoffice — and keep the final say with your editors.

Try it on your own Umbraco site: five single-page draft translation runs, with no time limit. AI provider usage costs are separate.