DiffusionGemma: 4x Faster Text Generation

Recently I had switched to OpenCode to try out many of the Non-US-Frontier-Labs models. My unexpected favorite model to use was Mercury (a diffusion model). Not because it was “smart” but because it was stupid fast. It was more of a pair-programming experience instead of the SOTA agentic experience of prompting and waiting. Honestly, it was also way more fun and brought back some of the pre-AI coding experience while still getting some benefits of AI. It felt less of a slot machine where you prompt, wait, and hope it went in the right direction. It made me even use the tiny models like Gemini Flash Lite and GPT Mini/Nano more too.

Anyways, so excited for an open-weight model and I hope it performs well. I’ll be testing this ASAP.

Some of these comments miss the advantage of diffusion. This is will have a big impact on edge devices, such as your phone or the GPU in your computer.

An LLM's decoder computes tokens one-at-a-time because attention has to account for each previous token. The existing LLM decoders scale well when you have enough load to batch many inferences together. Diffusion of limited benefit there. On edge you have a different problem: your inference accelerator is starved while sloshing GB of weights back and forth from RAM. That's because the consumer RAM like LPDDRx/GDDRx is lower bandwidth than HBM, and the requests are serial so you can't batch compute common weights. Diffusion can compute tokens in parallel which relieves the memory bandwidth bottle neck.

A good visual explanation of how text diffusion models like DiffusionGemma work: https://newsletter.maartengrootendorst.com/p/a-visual-guide-...

A few days ago I was just thinking that Google never talked about their diffusion text generation model after demoing it at I/O a year ago. The rumor is that it was too expensive to run, but with the provided chart using the same 1x H100 hardware and comparing DiffusionGemma to regular Gemma, that shouldn't be the case. I'm curious what the downside for this speed is here aside from being slightly weaker than Gemma.

What would a diffusing reasoning model look like? have a pre-defined length [thinking] block that gets diffused over a long time, and then the final output block uses what is in that thinking block as part of its input? And how do diffusion models decide the output length in the first place, is it a pre-set parameter? or does it diffuse an [end] token into the middle somewhere?

Can LoRAs be used to increase the quality of these diffusion models? Nvidia mentions something about this https://huggingface.co/nvidia/Nemotron-Labs-Diffusion-8B#inf...

I think this is the future. The sort of left-field rumble that turns into a quake in 5 years.

I just *love* the commit message on Github: "Make TPUs go brr"

It is cool but local models while okay already feel noticeably worse than even the cheapest APIs so I can't see myself sacrificing even a little bit of their quality for speed. I'm sure it's worth it for some usecases, curious to hear specific ones that people are already planning to deploy to production.

Is the diffusion approach any use in Multi-Token Prediction (MTP) drafters? https://blog.google/innovation-and-ai/technology/developers-...

We need more local open weight models that are performant and just as good (or good enough) as the best frontier ones.

Then you will be able to achieve Jevons Paradox and enjoy the same “productivity gains” without paying for these extortionate token prices by closed model providers or have it as cheap as possible.

And especially, no silent nerfing of the model.

[dead]