kohya-ss/controlnet-lllite

Pre-trained models and output samples of ControlNet-LLLite.

Note: The model structure is highly experimental and may be subject to change in the future.

Inference with ComfyUI: https://github.com/kohya-ss/ControlNet-LLLite-ComfyUI

For 1111's Web UI, sd-webui-controlnet extension supports ControlNet-LLLite.

Training: https://github.com/kohya-ss/sd-scripts/blob/sdxl/docs/trainllliteREADME.md

The recommended preprocessing for the blur model is Gaussian blur.

Naming Rules

controllllite_v01032064e_sdxl_blur_500-1000.safetensors

• —v01 : Version Flag.
• —032 : Dimensions of conditioning.
• —064 : Dimensions of control module.
• —sdxl : Base Model.
• —blur : The control method. anime means the LLLite model is trained on/with anime sdxl model and images.
• —500-1000 : (Optional) Timesteps for training. If this is 500-1000, please control only the first half step.

Models

Trained on sdxl base

• —controllllitev01032064esdxl_blur-500-1000.safetensors
• —trained with 3,919 generated images and Gaussian blur preprocessing.
• —controllllitev01032064esdxl_canny.safetensors
• —trained with 3,919 generated images and canny preprocessing.
• —controllllitev01032064esdxldepth500-1000.safetensors
• —trained with 3,919 generated images and MiDaS v3 - Large preprocessing.

Trained on anime model

The model ControlNet trained on is our custom model.

• —controllllitev01016032esdxlbluranime_beta.safetensors
• —beta version.
• —controllllitev01032064esdxlblur-anime500-1000.safetensors
• —trained with 2,836 generated images and Gaussian blur preprocessing.
• —controllllitev01032064esdxlcannyanime.safetensors
• —trained with 921 generated images and canny preprocessing.
• —controllllitev01008016esdxldepthanime.safetensors
• —trained with 1,433 generated images and MiDaS v3 - Large preprocessing.
• —controllllitev01032064esdxlfakescribble_anime.safetensors
• —trained with 921 generated images and PiDiNet preprocessing.
• —controllllitev01032064esdxlposeanime.safetensors
• —trained with 921 generated images and MMPose preprocessing.
• —controllllitev01032064esdxlposeanimev2500-1000.safetensors
• —trained with 1,415 generated images and MMPose preprocessing.
• —controllllitev01016032esdxlreplicateanime_0-500.safetensors
• —trained with 896 generated image pairs, 1024x1024 and 2048x2048 (highres. fix-ed).
• —Trained for 0-500 steps, but it seems to work for 0-1000.
• —controllllitev01032064esdxlreplicateanime_v2.safetensors
• —trained with 896 generated image pairs, 1024x1024 and 2048x2048 (highres. fix-ed).
• —Trained for 0-1000 steps.

About replicate model

• —Replicates the control image, mixed with the prompt, as possible as the model can.
• —No preprocessor is required. Also works for img2img.

Samples

sdxl base

[image]

[image]

[image]

anime model

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

[image]

replicate

[image]

[image]

Sample images are generated by custom model.