SDXL Lora Testing Workflow (ComfyUI) — v1.1

Use this ComfyUI workflow to test LoRAs with prompts for characters, poses, actions, LoRA combinations, styles, or concepts, mostly with SDXL models such as pony, illustrious, or base. It uses LoRA and scenario choices as inputs and lists a PreviewImage node for image output.

The workflow graph lists checkpoint and VAE loaders, text encoders, LoRA-stack controls, a sampler, a decoder, and an image-preview node.

Instructions are included in the workflow.

Listed node packs include comfy-image-saver, ComfyUI_ADV_CLIP_emb, ComfyUI_Comfyroll_CustomNodes, and comfyui_tinyterranodes.

Also listed are comfyui-easy-use, comfyui-image-saver, comfyui-impact-pack, and was-node-suite-comfyui.

Listed model files are Hyperrealistic_illustrious.safetensors, plantMilkModelSuite_walnut.safetensors, sdxlVAE_sdxlVAE.safetensors, and zy_illustrious_Realism_Enhancer_v1.safetensors.

For different tests, try varying characters, poses, actions, LoRA combinations, style, or concept.

Suggestion · not verified

In v1.1, the prompting process was remade to make it easier and more convenient.

Before running, confirm which installed ComfyUI pack matches was-node-suite-comfyui.

Suggestion · not verified

Check that the four listed model files are available under those exact names before use.

Suggestion · not verified