v25.1.2

Full Changelog: v25.1.1...v25.1.2

v25.1.1

What's Changed

• Style/harmonize colors by @bmaltais in #3253
• build(docker): upgrade to CUDA 12.8 and Debian 12 by @jim60105 in #3255

Full Changelog: v25.1.0...v25.1.1

v25.1.0

What's Changed

• Add --log_config support to GUI by @bmaltais in #2531
• Add support for custom learning rate scheduler type to the GUI by @bmaltais in #2532
• Add .webp image extension support to BLIP2 captioning. by @bmaltais in #2533
• Check for --debug flag for gui command-line args at startup by @b-fission in #2537
• Validate GPU ID accelerate input and return error when needed by @bmaltais in #2539
• Fix issue with pip upgrade by @bmaltais in #2540
• Adding the piecewise_constant LR Scheduler by @sangoi-exe in #2565
• PRevent providing the legacy learning_rate if unet or te learning rat… by @bmaltais in #2597
• fix: fixed docker-compose for passing models via volumes by @Retreatcost in #2601
• Fix toml noise offset parameters based on selected type by @bmaltais in #2602
• Fix adaptive_noise_scale value not properly loading from json config by @bmaltais in #2603
• Fix prompt.txt location by @bmaltais in #2606
• Improve "print command" output format by @bmaltais in #2607
• Use output model name as wandb run name if not provided by @bmaltais in #2612
• Bump docker/build-push-action from 5 to 6 by @dependabot in #2621
• Bump crate-ci/typos from 1.21.0 to 1.22.9 by @dependabot in #2620
• 173 error on train typeerror int argument must be a string a bytes like object or a real number not nonetype by @bmaltais in #2683
• Fix TE1/TE2 learning rate handling on dreambooth SDXL by @b-fission in #2697
• Auto-detect model type for safetensors files by @b-fission in #2690
• bring back SDXL Parameters accordion for dreambooth gui by @b-fission in #2694
• Update requirements_linux_docker.txt by @eftSharptooth in #2728
• Update IPEX and ROCm by @Disty0 in #2751
• Bump crate-ci/typos from 1.23.6 to 1.24.3 by @dependabot in #2765
• Add dark mode css variables by @wcole3 in #2802
• Update 3 files in kohya_gui folder by @iamrohitanshu in #2829
• Update requirements_pytorch_windows.txt by @wzgrx in #2848
• Bump crate-ci/typos from 1.23.6 to 1.25.0 by @dependabot in #2875
• Fix typo on README.md by @bulieme in #2876
• feat(docker): mount models folder as a volume by @vvsotnikov in #2868
• Specify custom tensorboard host via TENSORBOARD_HOST env var by @bmaltais in #2893
• Increase maximum Flux lora merge strength to 2 by @b-fission in #2915
• Fix typo by @bmaltais in #2927
• Update requirements onnxruntime-gpu by @b-fission in #2937
• Bump crate-ci/typos from 1.23.6 to 1.26.8 by @dependabot in #2940
• fix 'cached_download' from 'huggingface_hub' by @hqnicolas in #2947
• Bump crate-ci/typos from 1.23.6 to 1.28.1 by @dependabot in #2996
• Update README.md (for Runpod + Flux) by @ruucm in #3031
• Bump crate-ci/typos from 1.23.6 to 1.29.0 by @dependabot in #3029
• Update setup.sh by @CaledoniaProject in #3054
• Run on novita (#3119) by @bmaltais in #3120
• Bump crate-ci/typos from 1.23.6 to 1.30.0 by @dependabot in #3101
• updated prodigyopt by @Koronos in #3065
• fixed names on LR Scheduler dropdown by @Koronos in #3064
• Merge latest Sd3 flux.1 by @bmaltais in #3137
• Sd3 flux.1 uv by @bmaltais in #3051
• 3147 learning settings that were available in v2417 are no longer available in v2503 by @bmaltais in #3148
• Update lycoris_locon_extract.py, updated imports by @tdtrumble in #3184
• Apple Silicone Support (This time not on Master) by @rgeorgi in #3174
• Bump crate-ci/typos from 1.30.0 to 1.31.1 by @dependabot in #3153
• Update CUDA and PyTorch configuration in Dockerfile by @JoeYang1412 in #3195
• Bump crate-ci/typos from 1.31.1 to 1.31.2 by @dependabot in #3208
• Docs: Comprehensive README overhaul by @bmaltais in #3245
• build(docker): switch from pip to uv for package management by @jim60105 in #3229
• Refactor README.md to clarify uv vs pip usage by @bmaltais in #3246
• The message is fine as is. by @bmaltais in #3250
• docs: doc uv install method is fine for Docker use by @jim60105 in #3247
• Feat/add max grad norm dreambooth by @bmaltais in #3251
• v25.1.0 by @bmaltais in #3252

New Contributors

• @Retreatcost made their first contribution in #2601
• @eftSharptooth made their first contribution in #2728
• @wcole3 made their first contribution in #2802
• @wzgrx made their first contribution in #2848
• @bulieme made their first contribution in #2876
• @vvsotnikov made their first contribution in #2868
• @hqnicolas made their first contribution in #2947
• @ruucm made their first contribution in #3031
• @CaledoniaProject made their first contribution in #3054
• @Koronos made their first contribution in #3065
• @tdtrumble made their first contribution in #3184
• @rgeorgi made their first contribution in #3174
• @JoeYang1412 made their first contribution in #3195

Full Changelog: v25.0.3...v25.1.0

v25.0.3

• Upgrade Gradio, diffusers and huggingface-hub to latest release to fix issue with ASGI.
• Add a new method to setup and run the GUI. You will find two new script for both Windows (gui-uv.bat) and Linux (gui-uv.sh). With those scripts there is no need to run setup.bat or setup.sh anymore.

v25.0.2

• Force gradio to 5.14.0 or greater so it is updated.

v25.0.1

Fix issue with huggingface download issue

v25.0.0

v25.0.0 Release Notes

Major Features

• Flux.1 and SD3 Model Support: Added full support for Flux.1 and SD3 models, including GUI elements, parameters, LoRA training, and DreamBooth/finetuning capabilities.
• Advanced Learning Rate Options: Introduced custom learning rate scheduler support in the GUI, reworked TE1/TE2 learning rate handling for SDXL DreamBooth, and added lr_warmup_steps override.
• New Optimizers and Schedulers: Added support for AdEMAMix8bit, ScheduleFree, RAdamScheduleFree, ProdigyPlusScheduleFree, and Differential Output Preservation.
• LoRA Enhancements: Added LyCORIS LoRA support for Flux.1, increased maximum Flux-LoRA merge strength to 2, and introduced options for training specific blocks and swapping (blocks_to_swap).

Key Improvements

• GUI Enhancements: Implemented dark mode styles, improved visual sectioning of LoRA parameters, and added auto-detection of model types for safetensors files.
• Performance and Compatibility: Updated to the latest sd-scripts versions multiple times, improved requirements handling across platforms (Windows, Linux, Docker), and ensured compatibility with Python 3.10–3.12.
• Command and Output Improvements: Enhanced "print command" output format and used output model name as the default Weights & Biases (wandb) run name when not specified.

Bug Fixes

• Fixed pip upgrade issues and requirement validation problems.
• Corrected toml configuration issues for noise offset and adaptive_noise_scale.
• Resolved learning rate parameter syntax errors (text_encoder_lr, lr_warmup_steps, max_train_steps).
• Fixed various GUI issues, including Gradio on new installations, prompt.txt location, and SD3/Flux.1 LoRA training parameters (blocks_to_swap, fused_backward_pass).
• Addressed typos in README, GUI files, and logic errors in training scripts.

Detailed Changes

Feature Additions

• Added .webp image extension support for BLIP2 captioning.
• Introduced GUI support for Flux.1 and SD3, including parameters like guidance_scale, fp8_base, split_qkv, and cpu_offload_checkpointing.
• Added training support for T5-XXL with learning rate (t5xxl LR) and max token length fixes.
• Enhanced LoRA training with options like Rank for layers, uniform weighting scheme, and blocks_to_swap.
• Added startup options: --noverify to skip requirements validation and --skip_cache_check for the GUI.

Enhancements

• Updated sd-scripts submodule to the latest version multiple times, incorporating new features like fused_backward_pass for SD3 and pytorch-optimizer support.
• Improved Docker support by mounting the models folder as a volume and updating docker-compose.
• Added quiet output support for Linux setup and tensorboard host configuration via TENSORBOARD_HOST environment variable.
• Updated dependencies (e.g., torch, onnxruntime-gpu for CUDA 12, prodigyopt to 1.1.2) and requirements files for various platforms.

Dependency Updates

• Bumped crate-ci/typos from 1.21.0 to 1.30.0 across multiple updates.
• Updated docker/build-push-action from 5 to 6.
• Updated transformers, wandb, accelerate, IPEX, and ROCm for improved functionality.

Miscellaneous

• Updated README with a fork section and v25.0.0 release details.
• Fixed minor typos in GUI files (e.g., "Regularization") and README.
• Removed unsupported parameters from Flux LoRA network and reverted problematic max_train_steps changes.

For a full list of changes, refer to the commit history.

v24.1.7

• Fix issue with fastapi latest release causing gradio to fail. Updated to latest Gradio module release.

Full Changelog: v24.1.6...v24.1.7

v24.1.6

What's Changed

• Bump crate-ci/typos from 1.21.0 to 1.23.6 by @dependabot in #2677
• Fix issue with missing key. Upgrade Gradio release for security issue. by @bmaltais in #2682

Full Changelog: v24.1.5...v24.1.6

v24.1.5

What's Changed

• Security fix by @bmaltais in #2668

Full Changelog: v24.1.4...v24.1.5