【FLUX】LoRA TESTER — v1.0

【FLUX】LoRA TESTER — v1.0 is a ComfyUI workflow for testing LoRA settings with Flux.1 D: provide a model, text prompt, and LoRA settings to generate FLUX images, optionally upscale them, and save generation parameters with the images.

Load a model, write a prompt, adjust LoRA options, and generate text-to-image results.

Use the built-in SeedVR2 upscaler for tiled upscaling, with an on/off toggle.

Manual setup lists a base or GGUF model, CLIP files, text encoder files, and a VAE file.

Generated images are saved with generation parameters embedded for online publishing and remixing.

Before loading the workflow, enable ComfyUI’s “Use Nodes V2 (Vue)” setting and refresh the page.

Install the ComfyUI-UmeAiRT-Toolkit through ComfyUI Manager by searching “UmeAiRT,” or use the UmeAiRT Auto-Installer.

The listed model files are ae.safetensors, flux1-dev-Q8_0.gguf, t5-v1_1-xxl-encoder-Q8_0.gguf, and ViT-L-14-TEXT-detail-improved-hiT-GmP-TE-only-HF.safetensors.

The workflow lists these node types: DualCLIPLoaderGGUF, JWDatetimeString, Label (rgthree), Power Lora Loader (rgthree), SaveImage, and UnetLoaderGGUF.

Other listed node types are workflow/Image gen sampler, workflow/Other Nodes, workflow/Param, and workflow/positive prompt guidance.

Each 4x5 LoRA slot has an individual on/off toggle and strength control; additional LoRA modules can be connected.

The auto version downloads models automatically; manual setup uses the listed ComfyUI folders for diffusion models, CLIP files, UNet files, and the VAE.

Loading the workflow without “Use Nodes V2 (Vue)” enabled may cause display problems.