by XLabs-AI
Open source · 21k downloads · 422 likes
The Flux IP Adapter model is a specialized extension designed for Black Forest Labs' FLUX.1-dev model, enabling the integration of reference images into the image generation process. It supports resolutions of 512x512 and 1024x1024, providing flexibility for a variety of applications. This model excels at creating stylized images or those inspired by visual references, such as portraits or landscapes, by leveraging both textual prompts and input images. Its integration with ComfyUI simplifies its use for artists and developers looking to automate or customize their workflows. What sets it apart is its ability to balance visual precision with creativity while remaining accessible through user-friendly tools like ComfyUI's custom nodes.
This repository provides a IP-Adapter checkpoint for FLUX.1-dev model by Black Forest Labs
See our github for comfy ui workflows.
IP-Adapter is trained on 512x512 resolution for 50k steps and 1024x1024 for 25k steps resolution and works for both 512x512 and 1024x1024 resolution. Model is training, we release new checkpoints regularly, stay updated. We release v1 version - which can be used directly in ComfyUI!
Please, see our ComfyUI custom nodes installation guide
See examples of our models results below.
Also, some generation results with input images are provided in "Files and versions"
To try our models, you have 2 options:
git pull or reinstall it.model.safetensors from OpenAI VIT CLIP large, and put it to ComfyUI/models/clip_vision/*.ComfyUI/models/xlabs/ipadapters/*.Flux Load IPAdapter and Apply Flux IPAdapter nodes, choose right CLIP model and enjoy your genereations.If you get bad results, try to set true_gs=2
The IP Adapter is currently in beta. We do not guarantee that you will get a good result right away, it may take more attempts to get a result.
Our weights fall under the FLUX.1 [dev] Non-Commercial License