ComfyUI config (remote vast.ai)/lora/shaare/J7nlLA
git clone https://github.com/ltdrdata/ComfyUI-Manager
cd scripts
chmod +x install-comfyui-venv-linux.sh
./install-comfyui-venv-linux.sh
git clone https://github.com/ltdrdata/ComfyUI-Manager
cd scripts
chmod +x install-comfyui-venv-linux.sh
./install-comfyui-venv-linux.sh
Instant-LoRa 1.5 is a tool that improves the results of AI image manipulation, emphasizing the use of the standard IPAdapter ComfyUI. A notable feature includes a node that allows users to directly load images from a directory. Instant-LoRa collaborated with various projects, namely IPAdapter, ClipVision from Open AI, and ComfyUI, to enhance its functionality.
Users need to download various elements to utilize the tool: the IPAdapter ComfyUI, ComfyUI Impact Pack, ComfyUI Inspire Pack, Derfuu ComfyUI ModdedNodes, two more models ("ip-adapter-plus_sd15.bin" and "image_encoder/model.safetensors"), and ensure that their ComfyUI is up-to-date.
Using Instant-LoRa involves loading your choice of images and taking advantage of the tool's features. Recommendations include using varied, clear, and visually uncluttered images, setting the "Load vae" node to reflect your own value, and replacing the checkpoint with your own to prevent errors.
The quality of the results doesn't rely on training abilities but factors such as image selection, image clarity and quality, settings, base model, and prompt engineering. Users can achieve desired outcomes by replacing the images and adjusting values. The creator of Instant-LoRa vows to continue enhancing the tool while expressing excitement about the tool's potential.
Luigi Serafini is an Italian artist, architect and industrial designer, most famous for his book Codex Seraphinianus: an illustrated encyclopedia of an imaginary world, written in an imaginary language, with copious hand-drawn, colored-pencil illustrations of bizarre and fantastical flora, fauna, anatomies, fashions, and foods. which this LoRA is trying to imitate.
How to use:
Dimensions: 1024x1024 or the recommended dimensions for SDXL
Prompt: serafini style, a drawing of {subject}
Optional Negative prompt: Realistic, photography (although it works well without any)
Weight: 0.8 -1.2
serafini style
With this lora you'll be able to create several types of synths using these keywords. Some are followed by recommendations if you wish to get something closer to the original :
eurorack module
modular synth : add "cables, patch cables"
groovebox : add "colored touch pads, buttons"
volca
minilogue : add "keyboard, toggle switches, small display screen"
MS20 : add "keyboard"
TD-3 : silver, yellow and MO versions (you may have seen that i like acid)
TR-8O8 : add "colored buttons"
snth : a general keyword applied to all the dataset. Can be used for random results.
This Lora is captionned so try to always use these keywords in your prompts :
knobs, inputs, text, logos, LED indicators
Optional : keyboard, cables, patch cables, colored touch pads, buttons, small size, medium size, big size, wall size, (night, neon lights), man playing on a modular synth, studio, on table, multiple synths/modular synths
Lora weight : Anything works. I mostly used 0.8 or 1.
CFG scale: tested at 3 (on EpicRealism), Clip skip: 2
snth
Olafur Eliasson is an Icelandic-Danish artist known for his immersive installations, sculptures, and large-scale artworks that explore themes of perception, light, and nature. Born in 1967 in Copenhagen, Denmark, Eliasson's works often blur the boundaries between art, science, and environmental activism.
Eliasson's installations often incorporate natural elements, such as light, water, and air, to create immersive experiences that engage the senses and provoke contemplation. His use of natural materials and phenomena serves as a reminder of our connection to the environment and our impact on it.
Olafur Eliasson Style page
Yves Klein was a French artist who played a significant role in the development of post-war European art. Born in 1928, Klein is best known for his exploration of monochrome painting and his innovative use of a deep blue color known as "International Klein Blue" (IKB).
Klein's artistic practice was marked by his desire to transcend the limitations of traditional painting and to create an experience that went beyond the visual. He believed in the power of color as a spiritual force and sought to evoke emotions and sensations through his art.
One of his most iconic works is his series of monochromatic blue paintings, created using his patented IKB pigment. These works aimed to convey a sense of pure and infinite space, inviting the viewer to immerse themselves in the contemplation of color.
Yves Klein Style
Philippe Druillet is a French artist and writer known for his influential contributions to the realms of comics and graphic novels. Born in 1944, Druillet emerged as a prominent figure in the 1970s, and his distinctive artistic style left a lasting impact on the genre of science fiction and fantasy art.
Druillet's artwork is characterized by its dark, surreal, and intricate nature. He often depicted futuristic and dystopian worlds, combining elements of science fiction, fantasy, and cosmic imagery. His compositions are highly detailed, filled with intricate line work, and characterized by bold use of color and dramatic lighting.
Philippe Druillet Style
1024
Regularization images for SDXL 1.0 / Stable Diffusion 2.1 :
Install github-clone : pip install github-clone
then use :
ghclone https://github.com/tobecwb/stable-diffusion-Regularization-Images/tree/main/sd2.1/man
ghclone https://github.com/tobecwb/stable-diffusion-Regularization-Images/tree/main/sd2.1/woman
ghclone https://github.com/tobecwb/stable-diffusion-Regularization-Images/tree/main/sd2.1/person
or
ghclone https://github.com/yushan777/SD-Regularization-Images/tree/main/sdxl_person_ddim_1000
ghclone https://github.com/yushan777/SD-Regularization-Images/tree/main/sdxl_man_ddim_1000
ghclone https://github.com/yushan777/SD-Regularization-Images/tree/main/sdxl_woman_ddim_1000
ghclone https://github.com/yushan777/SD-Regularization-Images/tree/main/sdxl_artstyle_ddim_1000
ghclone https://github.com/hack-mans/Stable-Diffusion-Regularization-Images/tree/main/sdxl_artstyle_ddim
512x512
Regularization images for Stable Diffusion 1.5 :
ghclone https://github.com/tobecwb/stable-diffusion-Regularization-Images/tree/main/sd1.5/man
ghclone https://github.com/tobecwb/stable-diffusion-Regularization-Images/tree/main/sd1.5/woman
ghclone https://github.com/tobecwb/stable-diffusion-Regularization-Images/tree/main/sd1.5/person
ghclone https://github.com/aitrepreneur/SD-Regularization-Images-Style-Dreambooth/tree/main/style_ddim
ghclone https://github.com/djbielejeski/Stable-Diffusion-Regularization-Images-artstyle/tree/master/artstyle
Share and show ComfyUI workflow.
Example : https://comfy.icu/c/J7PoFyoP / https://comfy.icu/c/oWO9Dg
Easy tool to get PNG info, prompts, generation and model information of generated stable diffusion images.
Some tools add information about the parameters used for generating the image inside the png chunk. With PNGchunk.com you can easily get this information.
Load the image or image URL and within seconds you get the details of the generation.
ControlNet is an advanced image to image (img2img) tool that enhances the accuracy of reproducing poses and compositions from a base image with no need for excessive denoising. Unlike regular img2img tools, it doesn't leave traces of the base image in the final result. The tool uses a method of creating a 'control image' from the base image using edge detection, depth map detection, or pose identification. This control image is then employed during the image diffusion process. It also supports the usage of externally created maps or pre-posed skeletons, making it efficient and versatile.
Get the FP16.safetensors models for smaller filesizes and (presumably) less VRAM usage, get the .pth models if you want the original unpruned models or your card can't handle FP16.
Place either of these or all of them inside stable-diffusion-webui\extensions\sd-webui-controlnet\models
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_canny.pth
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_canny.yaml
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11f1p_sd15_depth.pth
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11f1p_sd15_depth.yaml
wget https://huggingface.co/lllyasviel/ControlNet/resolve/main/models/control_sd15_hed.pth
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_openpose.pth
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_openpose.yaml
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_scribble.pth
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_scribble.yaml
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_lineart.pth
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_lineart.yaml
Controlnet QR Code Monster v2 For SD-1.5 :
Instan LORA for SD-1.5 :
https://github.com/nerdyrodent/AVeryComfyNerd
wget https://huggingface.co/h94/IP-Adapter/resolve/main/models/ip-adapter-plus_sd15.bin
wget https://huggingface.co/h94/IP-Adapter/resolve/main/models/image_encoder/model.safetensors
SDXL ControlNet
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/OpenPoseXL2.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-canny-rank128.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-canny-rank256.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-depth-rank128.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-depth-rank256.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-recolor-rank128.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-recolor-rank256.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-sketch-rank128-metadata.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/control-lora-sketch-rank256.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/diffusers_xl_canny_full.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/diffusers_xl_canny_mid.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/diffusers_xl_canny_small.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/diffusers_xl_depth_full.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/diffusers_xl_depth_mid.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/diffusers_xl_depth_small.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/ioclab_sd15_recolor.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/ip-adapter_sd15.pth
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/ip-adapter_sd15_plus.pth
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/ip-adapter_xl.pth
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/kohya_controllllite_xl_canny_anime.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/kohya_controllllite_xl_depth_anime.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/kohya_controllllite_xl_openpose_anime.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/kohya_controllllite_xl_openpose_anime_v2.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/kohya_controllllite_xl_scribble_anime.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_canny_128lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_canny_256lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_depth_128lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_depth_256lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_recolor_128lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_recolor_256lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_sketch_128lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sai_xl_sketch_256lora.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sargezt_xl_depth.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sargezt_xl_depth_faid_vidit.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sargezt_xl_depth_zeed.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/sargezt_xl_softedge.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/t2i-adapter_xl_canny.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/t2i-adapter_xl_openpose.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/t2i-adapter_xl_sketch.safetensors
wget https://huggingface.co/ckpt/controlnet-sdxl-1.0/resolve/main/thibaud_xl_openpose.safetensors
Stable Diffusion XL :
wget https://huggingface.co/ckpt/sd_xl_refiner_1.0/resolve/main/sd_xl_refiner_1.0.safetensors
wget https://huggingface.co/stabilityai/sdxl-vae/blob/main/sdxl_vae.safetensors
SDXL support generating images at the following dimensions:
1024 x 1024
1152 x 896
896 x 1152
1216 x 832
832 x 1216
1344 x 768
768 x 1344
1536 x 640
640 x 1536
Stable Diffusion 1.5 :
wget https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned.safetensors