wxrks can power Augmented Translation Actions with its own built-in AI, or with your own OpenAI or Microsoft Azure OpenAI account. This article walks you through enabling the feature and, if you want to, connecting your own provider account.
​

💡 Who is this for? This guide is for Account Admins who need to enable Augmented Translation Actions and connect a custom OpenAI or Azure OpenAI provider within the wxrks platform.

1. In wxrks, click on Settings > Translation Settings > AI Actions.

2. Toggle on Enable Augmented Translation Actions.
​

3. The first time you turn it on, wxrks shows a Data Privacy notice explaining how your content is shared with the AI provider. Review it and confirm to continue.

By default, all operations use wxrks's own default models — no extra setup needed — but any operation can be routed to a different model, which overrides the wxrks model for that operation. If you'd rather use your own OpenAI or Microsoft Azure OpenAI account (for example, to use your organization's existing API credits or a specific deployed model), you can add it in Model Management.
​

⚠️ wxrks' integration with OpenAI and Azure uses the official OpenAI and Azure APIs directly, which are paid services. Make sure the account behind your API key has credits loaded.​
​

Note: adding a custom provider configuration is available starting on wxrks' FUSION plan. On other plans, this screen shows an upgrade prompt instead of the form described below.
​

1. Click on Settings > Account Settings > Model Management.

2. In the Advanced Model Management section, click Add Custom Setting. In the dialog that opens, choose the Scope (Account or a specific Org Unit), give the configuration a Name, and optionally limit it to specific Operations — leave this empty to use it as the default for any operation not explicitly covered.
​

Select OpenAI as the Provider. Before entering your key, make sure you generated it following the steps in How to create your Open AI API Key for the wxrks Integration. Then paste it into the API Key field and click Save. wxrks tests the credentials when you click Save. If the test fails, the row isn't saved and you see Could not validate followed by the provider, model and the provider's error message.

Select Azure OpenAI as the Provider. Before entering your details, make sure you generated them following the steps in Configuring wxrks Augmented Actions with Azure OpenAI Service. Then fill in the API Key, Resource Name, and Deployment Name fields and click Save. wxrks tests the credentials when you click Save. If the test fails, the row isn't saved and you see Could not validate followed by the provider, model and the provider's error message.

Once saved, wxrks uses your custom configuration for the operations you selected, within the scope you chose (the whole account or that Org Unit) — or as the default for any operation not otherwise covered, if you left Operations empty.​
​

Model Management supports OpenAI, Azure OpenAI, Anthropic, Groq, OpenRouter, Cerebras, Google Gemini (Studio) and Google Gemini (Vertex AI). See Configure Custom AI Providers with Model Management for all of them, and Connect your own AI engine for automatic post-editing (Context Sensitive) and quality evaluation (Smells) to route a specific operation to your own engine. For Gemini specifically, see Configuring wxrks Augmented Actions with Google Gemini.