| backend | ||
| embeddings | ||
| extensions | ||
| extensions-builtin | ||
| forgeChroma@dfefdea5dc | ||
| html | ||
| images | ||
| javascript | ||
| k_diffusion | ||
| localizations | ||
| models | ||
| modules | ||
| modules_forge | ||
| packages_3rdparty | ||
| RamTorch | ||
| scripts | ||
| _typos.toml | ||
| .eslintignore | ||
| .eslintrc.js | ||
| .git-blame-ignore-revs | ||
| .gitignore | ||
| .pylintrc | ||
| CHANGELOG.md | ||
| chromadct_optimization_integration.py | ||
| CITATION.cff | ||
| CODEOWNERS | ||
| convertpth.py | ||
| download_supported_configs.py | ||
| environment-wsl2.yaml | ||
| launch.py | ||
| LICENSE.txt | ||
| NEWS.md | ||
| package.json | ||
| pyproject.toml | ||
| README.md | ||
| requirements_versions.txt | ||
| sageattention-2.1.1+cu128torch2.7.0-cp310-cp310-win_amd64.whl | ||
| script.js | ||
| spaces.py | ||
| style.css | ||
| styles_integrated.csv | ||
| webui-macos-env.sh | ||
| webui-user.bat | ||
| webui-user.sh | ||
| webui.bat | ||
| webui.py | ||
| webui.sh | ||
Fork of forge to use Chroma!
This is a fork with the patch from https://github.com/croquelois/forgeChroma preinstalled thanks Thanks to @croquelois && @lllyasviel!
I am going to keep updating it with new features. Suggestions are welcome.
To use:
git clone https://github.com/maybleMyers/chromaforge
Download one of the t5xxl text encoder from https://huggingface.co/flux-safetensors/flux-safetensors/blob/main/t5xxl_fp16.safetensors and place in models/text_encoder.
Download the vae from https://huggingface.co/lodestones/Chroma/tree/main ae.safetenstors and put it in models/vae.
Download the main chroma model from https://huggingface.co/lodestones/Chroma1-HD/blob/main/Chroma1-HD.safetensors
You will need python3.10 installed.
Run webui-user.bat for windows installation. Run webui.sh for linux installation (./webui.sh) from the chromaforge root directory.
After you first run webui-user.bat it might error, if so, close it and re run it.
To update to the latest version navigate to your root directory in a terminal and type "git pull"
in Forge, on the top left select chroma
select the checkpoint and then in the next field select the text encoder and vae.
Use euler simple scheduler to start.
set the CFG scale to something like 3.5-7
forge doesn't seem to work with all quantized model, Q4_K_S fail, but Q4_1 work
refer to screenshot for working settings.
To update to torch 2.7.0 with cuda 12.8 on windows and install sage attention, navigate to your root directory after initial installation ie c:/chromaforge/ and run these commands:
venv/scripts/activate
pip install torch==2.7.0+cu128 torchvision==0.22.0+cu128 --index-url https://download.pytorch.org/whl/cu128
pip install -U "triton-windows<3.4"
pip install .\sageattention-2.1.1+cu128torch2.7.0-cp310-cp310-win_amd64.whl
there's a bunch of extra samplers/schedulers at these places:
https://github.com/DenOfEquity/webUI_ExtraSchedulers
https://github.com/MisterChief95/sd-forge-extra-samplers
RamForge
This supports use of the RamTorch memory management system. Add --use-ramtorch-chroma to your command line arguments. It will drastically lower vram usage to let you increase batch size or resolution greatly above what you could previously use. For instance you can run a batch of 8 1024x1024 images with a 12gb GPU. It does increas normal RAM usage by alot, 64GB recommended.
This was designed for training, it seems like not using it is faster.
https://github.com/lodestone-rock/RamTorch
Radiance Model
Download a checkpoint from here https://huggingface.co/lodestones/chroma-debug-development-only/tree/main/radiance then convert it with convertpth.py. Put it in your stable diffusion subfolder. Use the radiance model in the chroma tab of the gui without a vae. It will be automatically detected. Use the same T5 you normally use.
The memory management is currently derped. It will run fast with a 24GB card when setting the gpu weights to about 19,200, but if you need to offload expect 4x slowdown. You can dick with the gpu weights setting and it might run good if you set it to like half your vram.
Training: https://github.com/lodestone-rock/flow/ https://github.com/kohya-ss/sd-scripts/tree/sd3
Donate to Lodestone (training is bookoo expensive and crowdfunded): https://ko-fi.com/lodestonerock
Discord: http://discord.gg/SQVcWVbqKx
Changlog
9/24/2025
One trainer lora should load correctly now.
9/19/2025
Added support for RamTorch memory management system using normal chroma models. Use it by moidfying your webui-user.bat's set COMMANDLINE_ARGS= --use-ramtorch-chroma https://github.com/lodestone-rock/RamTorch
9/2/2025
Add res 2s, res 6s and res 16s samplers to main. Thanks to ClownsharkBatwing https://github.com/ClownsharkBatwing/RES4LYF
8/29/2025
Fixed previews w/ radiance chroma models, fixed memory management and offloading mechanisms for radiance somewhat, the memory management deffinately needs some more work.
8/28/2025
Added initial support for the radiance model class.
7/13/2025
Increase max res for chroma to 3072, add dedicated chroma button.
6/11/2025
Add links, Sigmoid Offset scheduler (thanks to croq and silveroxides https://github.com/silveroxides/ComfyUI_SigmoidOffsetScheduler), fix euler a simple sampler via croq's PR in forge https://github.com/lllyasviel/stable-diffusion-webui-forge/pull/2915.
5/27/2025
Fixed the sage attention implementation to work with chroma.
5/25/2025
Add support for sage and flash attention from this pr: https://github.com/lllyasviel/stable-diffusion-webui-forge/pull/2881 from @spawner1145
use the methods by adding --use-sage-attention or --use-flash-attention ... upon testing by a few people does not seem to have an increase on speed at all.
5/18/2025
Update to latest version of forge. Fix broken queuing.