Flux2 Klein Single/Dual Image Edit 9B Distilled with Lora Support — v2.0

This ComfyUI workflow edits one or two input images with LoRA support and outputs an edited image.

Choose one input image for a single-image edit or two input images for a dual-image edit.

Suggestion · non vérifié

Place flux-2-klein-9b-fp8.safetensors in ComfyUI/models/diffusion_models, qwen_3_8b_fp8mixed.safetensors in ComfyUI/models/text_encoders, and flux2-vae.safetensors in ComfyUI/models/vae.

Suggestion

Before downloading the model files, accept the agreement in BFL's repository.

Suggestion

The workflow names three model files: flux-2-klein-9b-fp8.safetensors, qwen_3_8b_fp8mixed.safetensors, and flux2-vae.safetensors.

rgthree-comfy version 1.0.2512112053 is required.

Version v2.0 replaces LoraLoaderModelOnly with Power Lora Loader (rgthree).

The listed base model is Flux.2 Klein 9B.

Verify that flux-2-klein-9b-fp8.safetensors is downloaded from https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-fp8.

Suggestion · non vérifié