this post was submitted on 08 Nov 2023
14 points (100.0% liked)

Stable Diffusion

5029 readers
2 users here now

Discuss matters related to our favourite AI Art generation technology

Also see

Other communities

founded 2 years ago
MODERATORS
db0
Even_Adder
14
GitHub - tencent-ailab/IP-Adapter: The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt. (github.com)
submitted 2 years ago by db0 to c/stable_diffusion
6 comments fedilink hide all child comments
 

This looks like a really cheap alternative to LoRas that anyone can use with little overhead

top 6 comments
sorted by: hot top controversial new old
[–] Even_Adder 3 points 2 years ago (1 children)

Pretty cool. I didn't know IP-Adapter could do all that.

[–] db0 3 points 2 years ago

Ye, it's kinda like a much versatile img2img

[–] zaknenou 1 points 2 years ago (1 children)

I feel like I will get roasted for asking but, is this possible with my laptop? :
RAM: 3.7 GiB
Processor: Intel® Core™ i5-4210U CPU @ 1.70GHz × 4
Graphics: Mesa Intel® HD Graphics 4400 (HSW GT2)
Nvidia GeForce 820M

[–] db0 1 points 2 years ago (1 children)

To run this particular mechanism, or Stable Diffusion in general?

[–] zaknenou 1 points 2 years ago* (last edited 2 years ago) (1 children)

umm, if you can explain what I can do with my hardware it would be cool. I actually have the idea that I might overheat my GPU or SSD if I try stable diffusion on it, or try some feature/ mechanism uncarefully.

[–] db0 2 points 2 years ago

you won't break anything by running stable diffusion, even if unsuccessfully. Just download ComfyUI and use low vram mode