bmaltais/kohya_ss

Issues

• Can we have a regularization weight like AI tookit reg_weight?

  #3083 opened a month ago by diodiogod
  1

• Add warning that extensions cannot be installed to Extensions > Available tab

  #3102 opened 12 days ago by tomasohara
  1

• train dreambooth by sd3.5 large cause error

  #3099 opened 13 days ago by iraymondking
  0

• Cannot Allocate Memory

  #3098 opened 13 days ago by tdashmike
  0

• Kohya GUI not passing accelerate arguments correctly

  #3097 opened 14 days ago by johnbrownlow
  0

• Stable Diffusion 3.5 missing parameters when generating toml file : Blocks to swap and Fused Backward Pass

  #3094 opened 18 days ago by FurkanGozukara
  7

• Is there any way to run kohya ss gui on RTX5090?

  #3096 opened 16 days ago by evoruta
  0

• Starting Gradio error on Kaggle : ERROR: Exception in ASGI application

  #3095 opened 17 days ago by FurkanGozukara
  1

• There is a capitalization error in the file kohya_gui/lora_gui.py

  #3093 opened 20 days ago by Ando-Lin
  0

• Install fails due to tensorflow==2.15.0.post1 requirement

  #3070 opened a month ago by StableLlama
  3

• module diffusers has no attribute FlowMatchEulerDiscreteScheduler. Did you mean: 'FlaxEulerDiscreteScheduler'

  #3091 opened 22 days ago by vedantroy
  0

• Kohya_SS GUI - LoRA Training Fails with 'NaN detected in latents' Error

  #3087 opened 23 days ago by 66Kratos
  5

• timm needs updating

  #3067 opened a month ago by AbstractEyes
  1

• Flux LoRA training crash randomly - `Segmentation fault (core dumped)`

  #3082 opened a month ago by ingTutui
  5

• subprocess.CalledProcessError: Command '['..Python311\\python.exe', '...kohya_ss/sd-scripts/train_network.py', '--config_file', 'C:/Users/yahowa/Desktop/thumbnail/dataset/trained_model/config_lora-20250215-161702.toml']' returned non-zero exit status 2.

  #3086 opened a month ago by MrYaho
  0

• Flex.1 Alpha LoRA/Finetuning

  #3056 opened 2 months ago by stepfunction83
  35

• Training stuck at epoch 1 steps 0

  #3081 opened a month ago by Red-Scarff
  1

• Can samples be generated with v-pred param?

  #3079 opened a month ago by Darudado
  0

• Does kohya work with an AMD 7900XTX on Win11?

  #3078 opened a month ago by Nugeorge
  0

• Adding a new optimizer

  #3077 opened a month ago by SushiLibanais
  0

• FLUX-Fill-LoRa-Training

  #3076 opened a month ago by FurkanGozukara
  1

• The virus I got from kohyass is long lasting and so annoying

  #3058 opened 2 months ago by 6Dii
  19

• Kohya SS raise subprocess.CalledProcessError

  #3075 opened a month ago by Bowlroller
  0

• ValueError: no metadata / メタデータファイルがありません: /metadata/meta_lat.json

  #3074 opened a month ago by BenDes21
  0

• Activate flux training ?

  #3073 opened a month ago by BenDes21
  8

• issue with accelerate

  #3072 opened a month ago by Abbyhub3
  2

• Unable to install CuDNN

  #3071 opened a month ago by OW1TY2
  1

• Runpod Template ?

  #3069 opened a month ago by BenDes21
  2

• error of use 2 and more GPUS for training FLUX model

  #3047 opened 2 months ago by NeuroDonu
  2

• Bug: max_grad_norm uses Lycoris value when Lycoris LoRA type not selected

  #3066 opened a month ago by stepfunction83
  0

• Schedule Free Optimizer availability in GUI

  #3061 opened a month ago by murphysez
  1

• Errors with creating LoRa

  #3057 opened 2 months ago by zetroxos
  1

• pics in folder/folder

  #3059 opened 2 months ago by Erlandsson
  2

• Installation broken at the moment for FLUX 1 branch

  #3060 opened a month ago by FurkanGozukara
  3

• Koyha pinokio fresh install but appaerance on chrome tiny, only 1/4 of full screen

  #3062 opened a month ago by eflin411
  1

• Setup script fails on Arch Linux: ModuleNotFoundError: No module named 'pkg_resources'

  #3050 opened 2 months ago by wonder75
  7

• how to disable guidance embeds

  #3055 opened 2 months ago by CodeAlexx
  0

• PDXL Lora Training Error

  #3052 opened 2 months ago by official-imvoiid
  4

• RuntimeError: cuDNN Frontend error

  #3049 opened 2 months ago by techidsk
  1

• LoRA Training Stuck with No Errors (Google Colab, CUDA 12.2)

  #3041 opened 2 months ago by Atsushi888
  11

• BLIP 2 caption fail with fresh install

  #3037 opened 2 months ago by RealArtGames
  2

• add MiGraphX dynamo backend for amd GPUs

  #3048 opened 2 months ago by Charmandrigo
  0

• Training failed exitcode : 1 (pid: 34080) / train_db.py FAILED (Multi-GPU failure)

  #3033 opened 2 months ago by Horang1173
  9

• I had a power outage while training, and now I am unable to train because of permission issues.

  #3045 opened 2 months ago by DalekSkaro1
  0

• i keep on getting voluptuous.error.MultipleInvalid: extra keys not allowed kept on deleting whatever the error shows it starts with bucketing i think so i removed it from toml it keeps on adding another error and now its saying voluptuous.error.MultipleInvalid: extra keys not allowed @ data['learning_rate']

  #3039 opened 2 months ago by DeFaker
  8

• FEATURE: lora merge with negative weight

  #3043 opened 2 months ago by kalle07
  0

• Therories about my failed trainings. Please Read.

  #3036 opened 2 months ago by Erlandsson
  1

• --network_train_unet_only How to add this command of the author? It always reports an error when added to Additional parameters

  #3040 opened 2 months ago by IAFFeng
  2

• Will intalling the cuda toolkit cause issues with my games?

  #3038 opened 2 months ago by Wendigo245
  1

• Stuck on setup.bat option 2 cudnn install

  #3032 opened 2 months ago by Horang1173
  1