Artificial Intelligence
Out of context: Reply #204
- Started
- Last post
- 1,374 Responses
- grafician0
"Running your own A.I. Image Generator with Latent-Diffusion
Run your own text to image prompts with CUDA, a bunch of disk space, and an insane amount of memory."
"In this post, we’ll look at getting setup with running your own A.I. image generator. You need a Linux system with a CUDA enabled card to get this working through to the end. I did get the basic txt-2img script working in Windows. Unfortunately, for the July released retrieval-augmented diffusion models, you need scann in order to index the openimages dataset, which is only available on Linux. WSL worked fine for me."
"Here’s roughly what I gather you’ll need:
Linux (WSL fine)
11.3 CUDA graphics
>= 32GB RAM
50GB disk space (~30GB without openimages)
3xxx series card (I am just guessing here. This is all being done with a 3090 FE)"