DE version is available. Content is displayed in original English for accuracy.
Advertisement
Advertisement
ā” Community Insights
Discussion Sentiment
85% Positive
Analyzed from 756 words in the discussion.
Trending Topics
#model#ball#target#base#output#more#tokens#faster#models#diffusion
Discussion (26 Comments)Read Original on HackerNews
And then through a LoRA adapter, you can ground the diffuser on the base modelās distribution (essentially have it ācompareā its proposals against what the base model wouldāve generated), which effectively means: exact same byte-for-byte output for the same seed, just roughly twice as fast (which should improve even more for batched tasks).
Iām not an expert, more of a āpracticing enthusiast,ā so I might be missing something, but at first glance, this reads super exciting to me.
So let's say a draft model generates 5 tokens, all 5 of these can be verified in parallel with a single forward pass of the target model. The target model may only accept the first 4 tokens (or whatever) but as long as the 5 forward passes of the draft model + 1 prefill of the target model is faster than 4 forward passes of the target, you will have a speedup while maintaining the exact output distribution as the target.
then once successfully trained you get faster inference from just the diffusion model
This startup seems to have been at it a while.
From our look into it - amazing speed, but challenges remain around time-to-first-token user experience and overall answer quality.
Can absolutely see this working if we can get the speed and accuracy up to that āgood enoughā position for cheaper models - or non-user facing async work.
One other question Iāve had is wondering if itās possible to actually set a huge amount of text to diffuse as the output - using a larger body to mechanically force greater levels of reasoning. Iām sure thereās some incredibly interesting research taking place in the big labs on this.
However quality is really important. I tried that site and clicked one of their examples, "create a javascript animation". Fast response, but while it starts like this
``` Below is a selfācontained HTML + CSS + JavaScript example that creates a simple, smooth animation: a colorful ball bounces around the browser window while leaving a fading trail behind it.
<!DOCTYPE html> <html lang="en"> <head> <meta charset="UTF-8"> <title>JavaScript Bounce Animation</title> <style> body, html { margin: 0; padding: 0;
```
the answer then degrades to
``` radius: BALL_RADIUS, color: BALL_COLOR, traivD O] // array of previous {x,y} positions }; ```
Then more things start creeping in
``` // 3ā£ Bounce off walls if (ball.G 0 ball.radius < 0 || ball.x + ball.radius > _7{nas.width) { ball.vx *= -1; ibSl.x = Math.max(ball.radius, Math.min(ball.x, canvbbF4idth - ball.radius)); } if
```
and the more it goes on the worse it gets
``` Ho7 J3 Works 0 Atep | Description | ```
and
``` ā¢ prwrZ8}E6on 5 jdF wVuJg Ar touc> 2ysteners ,2 Ppawn \?) balls w>SFu the 8b$] cliM#]9 ```
This is for the demo on the front page, so I expect this is a pretty good outcome compared to what else you might ask.
I also asked it some technical details about how diffusion LLMs could work and it provided grammatically-correct plausible answers in a very short time (I don't know the tech to say if it's correct or not).
https://www.emergentmind.com/topics/dflash-block-diffusion-f...
There are several Mac implementations of it that show > 2x faster Qwen3.5 already.
> 2025-04-12: Released I-DLM-8B, I-DLM-32B, and I-DLM-8B-LoRA on HuggingFace.
Is this old already? Not saying that's a bad thing, since it seems very sophisticated. Just curious if there's an update
https://huggingface.co/yifanyu/I-DLM-32B/tree/main
[0] https://docs.inceptionlabs.ai/get-started/models#mercury-2