Flux Dev LoRA Training Now Takes Just 30 Seconds

From 30 minutes to 30 seconds. Fal AI released a brand new model that gives Flux LoRA training a massive speed boost.

About six months ago, I trained a Flux image model using a small set of personal photos. It was part of an experiment to see how well the model could learn a subject and generate realistic results. The process wasn’t difficult, but it wasn’t exactly fast either — it took around 20 minutes to complete the training.

I loved the idea of creating a personalized image model. It became the main feature of my AI web app, Flux Labs AI.

At the time, I didn’t think much of it. That was just how long it took. But looking back now, it’s kind of wild to realize how far things have come in such a short time.

Fal AI just dropped a new Turbo Flux Trainer, and it’s ridiculously fast. My my initial experiments, I am getting the same kind of results, but the training process takes only 30 seconds.

In this guide, I’ll walk you through the full step-by-step process of training a Flux LoRA model in under a minute. I’ll also share some image examples so you can see just how good the output is. If you’ve ever wanted to create a custom image model based on your own photos or objects, now’s a perfect time to give it a go.

Let’s get started.

---

Step 1: Find the Turbo Trainer on Fal AI