Privately Host Your Own AI Image Generator With Stable Diffusion - Easy Tutorial!

Being able to self-host your own AI image generator free from interference is a powerful tool. Even though it might not quite reach the lofty heights of Dall-E and Midjourney out of the gate, you can train it and use pre-trained models to enchance your output. Learn how to host on Windows and Docker.
Easy Diffusion:
easydiffusion.github.io/
Docker Instructions:
github.com/AbdBarho/stable-di...
Recommended Hardware: github.com/JamesTurland/JimsG...
Discord: / discord
Twitter: / jimsgarage_
Reddit: / jims-garage
GitHub: github.com/JamesTurland/JimsG...
00:00 - Introduction to Stable Diffusion (Docker & Windows)
01:50 - Stable Diffusion - Windows
05:08 - Stable Diffusion - Docker
10:32 - Testing
12:10 - Outro

[@jimdavis3749]

Much appreciate the overall content and the fresh approach to home lab. You have successfully elevated my knowledge and appreciation all thing Jim's Garage !!

[@Jims-Garage]

Thanks, appreciate your feedback.

[@chrisumali9841]

Thanks for the demo and info, have a great day

[@Jims-Garage]

Thanks, you too.

[@happy9955]

good video thank you Jim

[@Jims-Garage]

Thanks 👍

[@RollerCoasterLineProductions]

Great content

[@Jims-Garage]

Thanks, a little bit different for the channel.

[@maciofacio364]

Thanks Jim for that, I have issues with VM/ubuntu/docker/Nvidia P4 as always ( I'm not a linux guy) but luckly there is a windows version which start working as soon as I installed this whole package everything works with Nvidia also. Thank you for that Jim once again.

[@Jims-Garage]

Good to hear, thanks for the feedback

[@mhavock]

Hi Jim! Did your install run fast enough on a CPU (your video shows about 4mins for the image)? have you tried the GPU install?

[@Jims-Garage]

Yes, it takes about 4min on CPU. About 15s on a 2080 Ti. I don't have an Nvidia GPU to test on my server unfortunately, and I'll likely run it locally if I ever need it (but to be honest Bing is just better...)

[@-rm-rf]

The robots are taking over

[@clivir8809]

?

[@basdfgwe]

More AI stuff!

[@Jims-Garage]

Last one for a while, promise :D

[@7lightcraft]

Will this work with apple M chips? I’m considering getting a high end mbp and I’d like to run SD if possible.

[@Jims-Garage]

Yes, I believe so: uxplanet.org/install-stable-diffusion-ui-on-mac-beginners-guide-351e40a9e8e2

[@SumNumber]

Cool. I will try this. Thanks for the share . Linux . :O)

[@Jims-Garage]

Linux is the second option

[@cyberspider78910]

Does this means if we go for Cloud hosted GPU solution, we can also install directly using Easy Diffusion ?

[@Jims-Garage]

Yes, I would think so.

[@cyberspider78910]

@@Jims-Garage Thanks...

[@sabra21]

lol, I tried the same prompt for warhammer 40k but bing refuses to generate the image because of "dangerous content" 😀

[@Jims-Garage]

Ha, it's a fickle thing. It'll probably work tomorrow 😂

[@inverse2k1]

@@Jims-Garagein my opinion it mostly depends on how you work your prompt. Talk around the obstacles, in short.

[@Jims-Garage]

@@inverse2k1 I agree, you can definitely iteratively refine the output and likely over time that will improve anyway

[@globalmodelsdotbiz8965]

Fooocus is all you need.

[@Jims-Garage]

Thanks, it's on my list

[@settlece]

when you get big and you will please dont just do the sponsor stuff you work is so good thanks so much
you make my home lab so much cooler

[@Jims-Garage]

If I ever get there I promise to use sponsored stuff to actually build homelabs, not just talk about products.