# New fast method for H3 is here, now we need a ComfyUI implementation — retained record

> Retained reddit record from reddit:StableDiffusion.

Canonical: https://highsignal.app/data/records/d022ea7fa741d587

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:StableDiffusion)  published 2026-09-07 17:27 ingested 2026-09-08 02:40

[original source ↗](https://www.reddit.com/r/StableDiffusion/comments/1w9yi3l/new_fast_method_for_h3_is_here_now_we_need_a/)

## retained text



```
[https://nvlabs.github.io/Sana/Sol-Engine/Sol-H3/](https://nvlabs.github.io/Sana/Sol-Engine/Sol-H3/)

This new fast method for H3 looks really interesting and now we need someone to bring this into ComfyUI.  
I dont know how well this will work on consumer cards, specially with lower VRAM, but I mostly use H3 on cloud GPUs so I will definitely try it there if we get a ComfyUI implementation.  
Really curious how much speed improvement we can actually get from this.

And wondering if we ever get the optimisation from the official team.

**Edit:** A dev has already jumped in to take a look at it and see if a ComfyUI implementation is possible. Really appreciate people like this in the community.

https://www.reddit.com/r/comfyui/s/QSM0KDFWKO
```



document fetched 2026-09-08 00:32 · canonical https://reddit.com/r/StableDiffusion/comments/1w9yi3l/new_fast_method_for_h3_is_here_now_we_need_a
