Giter VIP home page Giter VIP logo

Comments (8)

Foul-Tarnished avatar Foul-Tarnished commented on May 30, 2024 3

There's an app for android already
https://github.com/ShiftHackZ/Stable-Diffusion-Android

It doesn't use sd-turbo afaik, SD1.5 ?
Pretty slow, like 2min per image on SD865 12GB (+4GB) at 512x and something like 30 steps
I don't really remember

You can't even choose another checkpoint, and it lacks a lot of settings

from maid.

RookieIndieDev avatar RookieIndieDev commented on May 30, 2024 1

That's painfully slow. Mobile GPUs/CPUs aren't good enough I guess.

from maid.

GameOverFlowChart avatar GameOverFlowChart commented on May 30, 2024 1

There's an app for android already https://github.com/ShiftHackZ/Stable-Diffusion-Android
It doesn't use sd-turbo afaik, SD1.5 ? Pretty slow, like 2min per image on SD865 12GB (+4GB) at 512x and something like 30 steps I don't really remember
You can't even choose another checkpoint, and it lacks a lot of settings

thanks for sharing, it runs pretty decent, I ran until 320x448 20 steps in less than 40 seconds, after that resolution it crashed(probably not enough ram)

It doesn't show the option for steps to me but it shows 1/8? How do you know that it's 20?

Also yeah thanks for sharing I didn't know that the app can generate locally.

Edit: nvm it shows in the details of the finished image that it's 20 steps

from maid.

RookieIndieDev avatar RookieIndieDev commented on May 30, 2024

Certainly would be interesting addition. How well does it run on mobile devices locally?

from maid.

danemadsen avatar danemadsen commented on May 30, 2024

Certainly would be interesting addition. How well does it run on mobile devices locally?

No idea. Its theoretically possible because like llama.cpp its can be run in an android terminal emulator. Its going to be a huge addition once I get around to it and there's other stuff that needs to be done to core.cpp before I cross that bridge.

from maid.

Ar57m avatar Ar57m commented on May 30, 2024

Certainly would be interesting addition. How well does it run on mobile devices locally?

I ran sd-turbo safetensors q4_0 on stablediffusion. cpp about 88 seconds per iteration, it's too much slow

from maid.

MugoSquero avatar MugoSquero commented on May 30, 2024

You can try implementing whisper.cpp too. It runs quite easily on my phone using the base model.

from maid.

Ar57m avatar Ar57m commented on May 30, 2024

There's an app for android already https://github.com/ShiftHackZ/Stable-Diffusion-Android

It doesn't use sd-turbo afaik, SD1.5 ? Pretty slow, like 2min per image on SD865 12GB (+4GB) at 512x and something like 30 steps I don't really remember

You can't even choose another checkpoint, and it lacks a lot of settings

thanks for sharing, it runs pretty decent, I ran until 320x448 20 steps in less than 40 seconds, after that resolution it crashed(probably not enough ram)

from maid.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.