The best AI translation models aren’t the ones that perform well today. They’re the ones that perform best tomorrow.
Most teams approach AI translation the same way. Pick a model, run with it, and hope it holds up across every language. The problem is that no single model is best in every scenario. Different LLMs perform differently across languages, content types, and contexts. And the gap between the right model and the wrong one shows up exactly where you can’t afford it to: in your translation quality.
Smart Select was built to close that gap. Our enhanced version of Smart Select was built to further close that gap.
With the addition of OpenAI and Claude, Smart Select now taps into two of the most capable LLMs, expanding the AI engines it draws from when automatically selecting the highest-performing model for each language.
But what makes this Smart Select enhancement even more powerful is what happens behind the scenes.
Enhanced logging now captures and analyzes every translation decision Smart Select makes across languages. That intelligence continuously refines how models are selected. The more data Smart Select collects, the more accurate it becomes. And the more accurate it becomes, the better your translations get.
This isn’t a one-time upgrade. It’s a compounding one.
For teams scaling multilingual content across markets, that distinction matters. Translation quality that improves automatically, without oversight, manual testing, or ongoing configuration, is the kind of workflow advantage that compounds quietly and delivers loudly.
An even smarter Smart Select is here. And it’s only going to get smarter.
