# A small dataset unlocked general image editing in hours

Markdown copy of https://realalpha.app/s/a-small-dataset-unlocked-general-image-editing-in-hours/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: [Dustin Podell](https://realalpha.app/people/dustin-podell/) · Black Forest Labs
- Host: [Shawn Wang (swyx)](https://realalpha.app/people/swyx/) · Co-founder, AI Engineer · Wrote “The Rise of the AI Engineer” and founded Smol AI, maker of AI News.
- Show: Latent Space
- Episode: [Why FLUX 3 Trains on Images, Video and Sound — Andreas Blattmann & Dustin Podell, Black Forest Labs](https://realalpha.app/v/8cGfJOU3AlQ/), Nov 16, 2026
- Moment: 57:12 to 1:01:08 of the episode (4 min), playing on https://realalpha.app/s/a-small-dataset-unlocked-general-image-editing-in-hours/; the episode from there: https://realalpha.app/v/8cGfJOU3AlQ/?t=3432

## Quote

> “it was maybe like six hours of training, four hours of training. It was, it was a very, **very short training run**.”
> — [Dustin Podell](https://realalpha.app/people/dustin-podell/) at [59:28](https://realalpha.app/v/8cGfJOU3AlQ/?t=3568) · Latent Space

## Takeaways

1. A few carefully labeled image relationships sparked general editing.
2. Early behavior appeared after roughly four to six hours of training.
3. The team saw iterative editing as an alternative to prompt-and-regenerate.

## About this segment

Dustin Podell says a tiny set of high-quality image-relationship labels produced early editing behavior after a short training run. The result suggested the pretrained model already contained useful spatial knowledge.

From “Why FLUX 3 Trains on Images, Video and Sound — Andreas Blattmann & Dustin Podell, Black Forest Labs”, Nov 16, 2026. [Watch the full episode](https://realalpha.app/v/8cGfJOU3AlQ/)

## Transcript

From the transcript Latent Space publishes with the episode.

[57:18](https://realalpha.app/v/8cGfJOU3AlQ/?t=3438) So for me, like, so, so yeah, a- a- as Andy was saying, it originally started as a pitch of mine at some, uh, some cabin in France that, uh, I did essentially like a... I, I, I like to do these presentations every, you know, every offsite or every time we get together, where I try to look at, like, the state of the field and where I think the field is going, and kind of where... Like, what, what is something we're not doing or something where these models have a potential to do something novel and useful that, you know, maybe we're not looking at right now or maybe isn't solved.

[57:49](https://realalpha.app/v/8cGfJOU3AlQ/?t=3469) And so I actually did a presentation on, I know this is like the most overloaded word in the world at this point, but I was like, "Oh, it's the omni, the omni model presentation at that point," because, um, there was a lot of research work out there at the time on, you know, other editing models. If we remember back to, I think there was the old prompt to prompt, and then there's, uh... Why, why can I not remember the name of this, um...

[58:09](https://realalpha.app/v/8cGfJOU3AlQ/?t=3489) The pitch was essentially these models are clearly learning, like, these deep relationships about the world, and this is gonna keep coming up. It's about these, these relationships. And to me, you know, like, the relationship of like, okay, this Coke can is sitting on top of a table. This bottle is sitting here. And you prompt the model, and you go, "Okay, I want a photo of a dog wearing a hat." And it needs to know, okay, the hat is on top of the dog. It, it, it needs to have this general understanding.

[58:30](https://realalpha.app/v/8cGfJOU3AlQ/?t=3510) And my kind of theory in mind at the time was that since the model already has all of this, like, spatial knowledge, if we essentially build the right data and train the model, like train a, a, an already pre-trained model the right way, we should be able to evoke essentially, like, general image editing with very, very little training [laughs], capturing these already, like, pre-built relationships of when things are, are, are there, when they're not there, uh, and how these, like, relationships, these, these space relationships inside of an image work.

[58:58](https://realalpha.app/v/8cGfJOU3AlQ/?t=3538) And then essentially what it became is like a three to form. I don't remember the timeline of, uh, me here over in our SF office with the one data guy I stole, building out essentially just early test training data sets. And, and I remember the moment I saw that this was, like, clearly working quite well when we made this effectively tiny data set of just very, like, high quality labeled. It was essentially describing, like, simple relationships in an image, and that was really it. Uh, like, it wasn't a way that you would prompt a, a modern, like, edit instruct model.

[59:30](https://realalpha.app/v/8cGfJOU3AlQ/?t=3570) It was just literally describing these relationships. And I trained the model for, like, I mean, it was maybe like six hours of training, four hours of training. It was, it was a very, very short training run. And all of a sudden, like, I saw what felt like a clear generality coming out of it, 'cause it was-- I think it was, like, a photography data set. It was just, like, photos of stuff. But even just from that, I was seeing, like, I can put a logo on a shirt. I can put, like, I can do... And it wasn't good.

[59:55](https://realalpha.app/v/8cGfJOU3AlQ/?t=3595) It was just, like, the model was clearly learning to attend between these two concepts and use these pre-trained relationships. And then I took that to our next offsite, showed results, and then it became this essentially more company-wide project for about, you know, um, a, a couple months, where we really put effort in, brought people in. You know, really improved essentially... I, I, it was mostly a data set project. Like I'll say, I'll say it was mostly a data set project. And then in the end of it, the distillation.

[1:00:19](https://realalpha.app/v/8cGfJOU3AlQ/?t=3619) Um, and then there, there was a lot of, there was a lot of very nice distillation work by Axel that was done to, to, to speed it up. Then we essentially were able to put out, put this model out into the world, and I, I was quite proud of. Yeah. And- That was the first kind of model to really do this character consistent editing, right? In, in, like, high speed. Yeah. That made it, again, accessible to so many people. For most people, it's very common to generate something, chat with it, change it.

[1:00:42](https://realalpha.app/v/8cGfJOU3AlQ/?t=3642) Before this, image generation took quite a while, and you need to recreate the image. You couldn't- Yeah... change the parent to be blue. Just, just getting the same character, it won't- It wouldn't be the same. You- Or you need to mask, uh, something that you don't like- Yeah... and, and then fill it in the blank. And then- Yeah, I just, I wanted to get away from the slot machine, essentially. Everyone had to do the slot machine. [laughs] Yes. Like prompt and prompt and prompt and prompt, and it was very clear that, like, iterative editing was, like- Oh, no... much more effective way to do this.
