Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updating Flux Quickstart Doc with Pre-Trained Model Info #1116

Merged
merged 2 commits into from
Nov 2, 2024
Merged

Updating Flux Quickstart Doc with Pre-Trained Model Info #1116

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d32b47c
Updating Flux Quickstart with configuration information for pre-train…
riffmaster-2001 Nov 2, 2024
aa31083
Update documentation/quickstart/FLUX.md
bghira Nov 2, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 15 additions & 0 deletions documentation/quickstart/FLUX.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -515,6 +515,21 @@ When you do these things (among others), some square grid artifacts **may** begi
- However, if you're looking to improve results equally across many aspect buckets, you might have to experiment with `crop_aspect=random` which comes with its own downsides.
- Mixing dataset configurations by defining your image directory dataset multiple times has produced really good results and a nicely generalised model.

### Training custom fine-tuned Flux models

Some fine-tuned Flux models on Hugging Face Hub (such as Dev2Pro) lack the full directory structure, requiring these specific options be set.

Make sure to set these options `flux_guidance_value`, `validation_guidance_real` and `flux_attention_masked_training` according to the way the creator did as well if that information is available.
```json
{
"model_family": "flux",
"pretrained_model_name_or_path": "black-forest-labs/FLUX.1-dev",
"pretrained_transformer_model_name_or_path": "ashen0209/Flux-Dev2Pro",
"pretrained_vae_model_name_or_path": "black-forest-labs/FLUX.1-dev",
"pretrained_transformer_subfolder": "none",
}
```

## Credits

The users of [Terminus Research](https://huggingface.co/terminusresearch) who worked on this probably more than their day jobs to figure it out
Expand Down
Loading