The AI Breakthrough That Uses Almost No Power to Create Images

Posted4 months agoActive4 months ago

Luc

12 points

5 comments

techxplore.comTechstory

excitedpositive

Debate

20/100

AIEnergy EfficiencyImage Generation

Key topics

Energy Efficiency

Image Generation

A new AI breakthrough uses a diffusion-based image generator that significantly reduces power consumption, sparking interest and discussion among HN users about its potential applications and technical details.

Snapshot generated from the HN discussion

Discussion Activity

Light discussion

First comment

Peak period

2-3h

Avg / period

1.7

Key moments

01Story posted
Sep 4, 2025 at 8:12 AM EDT
4 months ago
Step 01
02First comment
Sep 4, 2025 at 10:14 AM EDT
2h after posting
Step 02
03Peak activity
2 comments in 2-3h
Hottest window of the conversation
Step 03
04Latest activity
Sep 4, 2025 at 1:27 PM EDT
4 months ago
Step 04

Generating AI Summary...

Analyzing up to 500 comments to identify key contributors and discussion patterns

Discussion (5 comments)

Showing 5 comments

kbelder

4 months ago

1 reply

"The new diffusion-based image generator works by first using a digital encoder (that has been trained on publicly available datasets) to create the static that will ultimately make the picture. This requires a small amount of energy. Then, a liquid crystal screen known as a spatial light modulator (SLM) imprints this pattern onto a laser beam. The beam is then passed through a second decoding SLM, which turns the pattern in the laser into the final image."

Ok, I'm going to go read the paper now, but I have to say, this sounds like absolute nonsense.

kbelder

4 months ago

1 reply

Ok, I don't understand all of the paper. I do like that fact that it's combining methods from physics and computer science, because there are probably many more discoveries to be made in those kinds of permutations.

But it seems, if I'm interpreting the paper correctly, that the real gain is that there is a static decoder filter that is generated for one particular prompt, and then it may be reused over and over at minimal cost, to generate butterfly after butterfly. Transforming the diffuser from a general-purpose generator to a specialized, more-efficient one that needs to be rebuilt if a different type of image is desired.

dekhn

4 months ago

To me the simplest place to start is Fourier Optics. A carefully set up collection of lenses, illuminators, samples, and cameras can compute (well, physically evaluate) the fourier transorm of a 2D image. https://en.wikipedia.org/wiki/Fourier_optics (so you can for example do convolutions)

From what I can see in this area, the physical set up mainly computes forward passes by passing light through the SLM, while the computer computes the backward pass (using a mathematical equation that approximates the forward optics and computes its derivative). The nice thing is that SLMs are simply programmed to have a specific intensity pattern, it doesn't need a whole manufacturing pass to update the SLM.

majidmir

4 months ago

Original paper: https://www.nature.com/articles/s41586-025-09446-5

tanseydavid

4 months ago

While I do not fully understand how it works, this sounds amazing and should be applicable (eventually) to other workloads.

View full discussion on Hacker News

ID: 45126362Type: storyLast synced: 11/20/2025, 11:26:10 AM

Want the full context?