In this article you will find information on how to use a different AI translation engine, by adding your own API key, manage the settings that this unlocks, as well as information of the global AI Translation Settings that work for all translation engines.

Let's start with the AI Translation management first and it's settings.

AI translation engine management (available on the Premium plan) allows you:

To add your own API key, go to the Settings > AI Translation Settings page

Choose the desired engine, then add our API key, and click "Verify key" button on the right:

If you get a "Verified" success message, you are good to go - future AI translations will use the engine you have selected. Otherwise, if the verificaiton fails, please make sure:

Some of the engines have specific settings that you can use for further customization of the transalations.

Using OpenAI API key allows you to:

Using Anthropic API key allows you to:

Adding your DeepL API key allows you to configure:

Adding DeepSeek API key allows you to choose:

Beshides choosing the AI engine by adding your own API key, there are other AI settings on the Settings > AI Translation Settings page, that are global for all translation engines.

By default, you receive an email notification once a bulk AI Translation task is complete. You may disable those notifications by disabling the "Send email notification after AI translate" checkbox:

This tells the AI whether it should try to guess your source language from which you are translating, or to explicitly use the primary language of your store as a source language.

The first option works for the majority of the stores. But let's say you have product names that have words from multiple languages - for example, your store primary language is spanish but the product names also contain a brand name in English. In this case, you may choose "Use store default language" for this setting, so that the Spanish words are translated and the English words are skipped.

The handle translations are disabled by default to prevent 404 errors. The only way to unlock the handle bulk translation is to enable the "Enable handle translation" checkbox, which is available on all paid plans:

This is only recommended if you have technical knowledge and can handle preventing any potential broken links.

By default, bulk AI Translations process only the Active products in your store. On this page, you may allow the bulk AI translations of Draft, Unlisted and/or Archived products if you need it:

Store context is a short description of your store: what you sell, who you sell to, and your industry. Translation Lab sends it with every AI translation, so the engine knows the subject area of your content. Words with several meanings then get the right one, and product terms match your field.

Store context describes your store. It does not give orders. To control tone or style, such as formal or informal address, you can add your own API key and Special instructions. Your glossary and formatting rules always come first, and store context cannot override them.

With the default Translation Lab engine, you can enter up to 300 characters. If you connect your own API key, there is no character limit, but note that longer context consumes more tokens and may affect your billing.

A store sells fly-fishing equipment and enters this store context:

We sell fly-fishing equipment: rods, reels, flies, lines and tackle for anglers of all levels.

Without store context, the engine might translate "fly" as the insect or "tackle" as the sports move. With store context, it uses the fishing terms in each language. So "Dry flies – 12 pack" becomes "Trockenfliegen – 12er-Pack" in German.

Question: Which is the best translation engine - OpenAI, DeepL, DeepSeek, Anthropic or Google Translate?

Answer: There is no universally “best” translation engine.

Quality and cost vary depending on:

Some engines perform better for European languages, others may handle Asian languages differently. Pricing structures also differ - some charge strictly per character, others per token or usage tier.

The most reliable approach is to test multiple engines with a few of your products (not all the content), then compare quality, consistency, and cost for your specific use case.

Question: How much would the translations cost be with each engine?

Answer: Pricing structures differs - some translation engines charge strictly per character, others per token or usage tier. It's best to refer to the selected engine's pricing page for reference.

Since it's impossible to calculate the exact amount of words/characters a given store has, the best way to get an idea is to translate 10 products, for example, then check your billing at the selected platform (note that some of them take a few hours to update your usage statistics)

Question: Why would I want to add my own API key?

Answer: There are 2 reasons why you might want to use your own API key:

Question: Which engines work without an API key?

Answer: Only the Default engine. OpenAI, DeepL, DeepSeek, Anthropic and Google Translate all require your own API key.

Question: What model does the Default engine use?

Answer: Our Default engine is based on the OpenAI ChatGPT model.

Question: What are the "Special instructions" I can add to the settings of some AI engines?

Answer: Special instructions are simply additional "prompt" (or could be considered an AI "skill") you can provide in the settings when you add your own API key for some engines.

Question: I'm using my own API key and my AI Translations task was interrupted because my API credit ran out. Will the task continue once I add credit?

Answer: No, the task won't resume automatically. If a one-time AI Translations task ends with an "Interrupted" status because your API key ran out of credit, you'll need to:

When you connect your own API key, all translation usage is billed by your AI provider, not by TLab. There is nothing to top up or pay for inside the app.