Tensor.Art Green
Create

This is the list of helpful articles (on going)

Updated:

1. Resource List: Characters in FLUX

By kftiger at Civitai

Resource List: Characters in FLUX

This is a dynamic list of characters that already exist in the FLUX base dataset that do not require any additional LORAs to produce, only needing to be named in your prompts.

Example from Real People:

Albert Einstein, Angela Lansbury (Young), Angelina Jolie, Arnold Schwarzenegger, Barack Obama, Bill Clinton, Bruce Lee, Bruiser Brody, Colin Baker, Daniel Craig, DJ Marshmello, Donald Trump, Dwayne Johnson, Emma Stone, Fairuza Balk, George W Bush, Hulk Hogan, Jackie Chan (50/50), Joe Biden, John Pertwee, Johnny Depp, Lady Gaga, Mark Zuckerberg, Meryl Streep, Michael Jackson, Queen Elizabeth, Stan Hansen (50/50), Stan Lee, Will Smith, Wolfgang Amadeus Mozart.

found by myself

  • Naruto Uzumaki

  • Sasuke Uchiha

  • Kakashi Hatake

  • Power Ranger

  • Kim Jong-un

2. 480+ Pony Diffusion XL Hats, Masks and More Props List - Booru Tags (SFW)

by Viola_Aiarty at Civitai

480+ Pony Diffusion XL Hats, Masks and More Props List - Booru Tags (SFW)

This post belongs to my series of Pony Diffusion XL Recognized Tokens lists. You can use the quick navigations below to access previous posts.

for more detail go to the post or this one Access all the original image grids and wildcards here: https://mega.nz/folder/XcVVXbRL#uQPsHUfLeBnaNhSTmDgziQ

3. How to train a logo lora for Flux

by JaneB at Civitai

How to train a logo lora for Flux

from this article, i made test on tensor.art lora training. you can play it here

https://tensor.art/models/780240303475923311/Persib-logo-E9

https://tensor.art/models/780138126203986771/Gojek-Logo-E4

4. Flux model training from just 1 image [Attention Masking]

by mnemic at Civitai

Flux model training from just 1 image [Attention Masking]

this is article guide to train lora with single image and use masking, but no option for masking at tensor.art training, so if you train with transparant PNG background, it will failed.

mnemic also have good article about captioning etc, so this is the bunch of his article

  • https://civitai.com/articles/7146/flux-style-captioning-differences-pt2-4-new-caption-tools-training-diary

  • https://civitai.com/articles/7044/font-model-training-with-flux-training-diary (for more detail about font, here https://civitai.com/user/juaner0211589/)

  • https://civitai.com/articles/6868/flux-character-caption-differences-training-diary

  • https://civitai.com/articles/6792/flux-style-captioning-differences-training-diary

  • https://civitai.com/articles/2138/lora-datasets-training-data-list-civitai-dataset-guide

5. Dont training lora with flux 1 dev.

this is article is interesting, to train lora just used finetuned model instead the original from black forest lab.

https://medium.com/@zhiwangshi28/why-flux-lora-so-hard-to-train-and-how-to-overcome-it-a0c70bc59eaf

i will update post soon. thanks



11
0

Comments