The Working Wonders blog

Understanding nvision Training Times: Data, Hardware, and Engineering Trade-offs

One of the most common questions around AI-based surrogate modelling is simple:
How long will model training take?

The honest answer is that there is no universal number. Training duration depends on many factors, including dataset size, number of outputs, hardware configuration, and model complexity. However, understanding how these factors influence training behaviour is often more useful than trying to estimate absolute durations.

What’s New in nvision 2026.1

n engineering, more is not always better. More epochs, more compute… it is easy to overcommit upfront to be on the safe side. But what if model training could adapt to your needs instead of forcing you to guess them?

With nvision 2026.1, we introduce features that make training more flexible, more economical, and better aligned with how engineers actually work.

How AI Augments Engineering Thinking | Data-Driven Surrogate Models Explained

AI has been transforming engineering – but, what can engineers (and cannot) expect from AI models.