Skip to content

Issues: kohya-ss/sd-scripts

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

656 Open 539 Closed
656 Open 539 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

how to set timestep shift when train flux-dev lora?
#1907 opened Jan 28, 2025 by SlZeroth
1
Shouldn't fix_noise_scheduler_betas_for_zero_terminal_snr be applied BEFORE prepare_scheduler_for_custom_training? help wanted Extra attention is needed
#1905 opened Jan 27, 2025 by 67372a
1
Flux training, no matter the settings I get: returned non-zero exit status 3221225477
#1902 opened Jan 27, 2025 by alexgilseg
1
Implement Test and Validation Set Loss
#1897 opened Jan 24, 2025 by stepfunction83
2
Does the full fine-tuning in SD3.5 have a setting similar to --guidance_scale?
#1896 opened Jan 24, 2025 by raindrop313
lora extracted from dreambooth trained model has poor compatibility with controlnet
#1892 opened Jan 23, 2025 by Nomination-NRB
Add Support for Flex.1 Alpha
#1891 opened Jan 22, 2025 by stepfunction83
21
Diffusion Feature Extractor ostris/ai-toolkit implemented
#1890 opened Jan 22, 2025 by FurkanGozukara
Have any guide for training flux Fill lora, or fine-turn flux Fill?
#1889 opened Jan 21, 2025 by CaiNiaoCYY
2
Insufficient RAM to load Models when using sdxl_gen_img.py in Free Google Colab Environment
#1887 opened Jan 20, 2025 by DKnight54
flux_models.py:85: UserWarning: 1Torch was not compiled with flash attention.
#1884 opened Jan 19, 2025 by itasky
Encountering _IncompatibleKeys while training LoRA for flux
#1880 opened Jan 18, 2025 by xcltql666
SDXL Training prompt length limited to 225 tokens or can be more?
#1877 opened Jan 15, 2025 by onirhakin
学習させてもsafetensorデータが生成されない
#1874 opened Jan 13, 2025 by m0r1s1
3
RAdamSchedulefree give great results but errors after 1 epoch
#1872 opened Jan 10, 2025 by Deathawaits4
flux_train.py when deepspeed enable RuntimeError: mat1 and mat2 must have the same dtype, but got Float and BFloat16
#1871 opened Jan 10, 2025 by mobilejammer
Huber loss not supported??
#1868 opened Jan 7, 2025 by Deathawaits4
3
How to train SDXL LoRA on specific blocks?
#1863 opened Jan 2, 2025 by rishipandey125
args.noise_offset is not working for flux training？
#1861 opened Dec 31, 2024 by ziyaxuanyi
2
Text Encoders & Inference - SD3.5L Dreambooth 💫
#1860 opened Dec 30, 2024 by deman311
4 tasks
torch 2.4.0 does not work with fp8 training
#1859 opened Dec 27, 2024 by ddpasa
1
Failed to offload sd3.5L to cpu
#1857 opened Dec 27, 2024 by moonlightian
Flux.1-dev lora training not support trigger word parameter？
#1854 opened Dec 25, 2024 by Nine98
1
Feature Request - Differential Output Preservation loss For Dreambooth/Finetuning
#1852 opened Dec 24, 2024 by XCogni
Flux Controlnet Error while processing rearrange-reduction
#1849 opened Dec 21, 2024 by WainWong
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.