ComfyUI Image Workflows — V25

ComfyUI Image Workflows — V25 is for generating images from text prompts and detailing existing or previously generated images with XL, Illustrious, and NoobAI models; it takes prompts or images and produces generated or detailed images.

Advanced is described as a comprehensive text-to-image workflow; Standard is a simpler Advanced version, Basic is a simpler Standard version, and Detailer works on existing or previously generated images.

Named prompting and generation features include LoRA, TIPO Prompting, Detail Enhancers, HiresFix, Wildcards, and Regional Prompting.

Other listed image features are ControlNet Upscale, IPAdapter, OpenPose, ControlNET, Color Match, FBCNN, Inpaint, and Outpaint.

The listing does not specify a model-use license.

The listed ComfyUI node packs are comfyui_controlnet_aux, comfyui_ipadapter_plus, comfyui_ultimatesdupscale, comfyui-custom-scripts, comfyui-easy-use, comfyui-fbcnn, comfyui-image-saver, comfyui-impact-pack, comfyui-impact-subpack, comfyui-lora-manager, comfyui-ppm, rgthree-comfy

Other named node packs are z-tipo-extension. Named model files include 4x_foolhardy_Remacri.pth, bbox/Eyeful_v2-Paired.pt, bbox/face_yolov9c.pt, bbox/hand_yolov9c.pt, bbox/yolo11m.pt, clip_vision_g.safetensors, CLIP-ViT-bigG-14-laion2B-39B-b160k.safetensors, CLIP-ViT-H-14-laion2B

Other named model files are control-lora-canny-rank256.safetensors, fabricatedXL_v50.safetensors, fabricatedXL_v60.safetensors, ip-adapter-faceid-plusv2_sdxl.bin, noobaiInpainting_v10.safetensors, noobaiXLControlnet_openposeModel.safetensors, sam_vit_b_01ec64.pth, SDXL/sdxl_vae.s

V25 notes say default values were adjusted and CFGZeroStar was added to text-to-image.

The notes say prompt and conditioning handling was made less obfuscated and LoRA Manager was added to Basic.

Before choosing a workflow, check that its named node packs and model files are available in your ComfyUI setup.

Vorschlag · nicht geprüft

The V25 update notes say they are not an exhaustive changelog.