How to use embeddings through API #13402
Unanswered
SebastiaanSG
asked this question in
Q&A
Replies: 1 comment
-
It's not the embeddings. You forgot the underscore in And it already has metadata (otherwise for i in r['images']:
with open(output_file, 'wb') as f:
f.write(base64.b64decode(i)) If you're only doing one image at a time, you don't need a loop. with open(output_file, 'wb') as f:
f.write(base64.b64decode(r['images'][0])) |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I'm using the api to generate images through python, but when I use the same generation data in the WebUI the images are way prettier. Then I found out it was because my (negative) embeddings aren't working.
I'm using:
negativeXL_D unaestheticXLv13
But they both aren't registering when I just use them as text:
payload = {
"prompt": input_prompt,
"steps": 50,
"cfg_scale": 7,
"negative prompt": "negativeXL_D unaestheticXLv13, text, words, letters, signature,"+additionalnegative,
"width": imgwidth,
"height": imgheight,
"sampler_index": "Euler a",
}
This is part of the function I'm using for this.
Can anyone help or explain to me how I can get the negative embeddings to start working,
Thanks in Advance!
Sebastiaan
This is the entire script, should it be helpful:
Beta Was this translation helpful? Give feedback.
All reactions