LucidFlux

W2GenAI-Lab/LucidFlux is ranked 10,791 among these 8 peers; OpenBuddy/OpenBuddy is ranked 10,796.

26]** - LucidFlux now integrates PiD to enable **4K image restoration**! 19]** - We released the training code for LucidFlux. 13]** - LucidFlux now integrates UltraFlux's VAE to enable **2K image restoration**! 10]** - We released the metadata for the clean images used in LucidFlux at LucidFlux-Training-Data and the filtering pipeline in . 06]** - LucidFlux is accepted by **ICLR'26**. LucidFlux is a caption-free universal image restoration framework that leverages a lightweight dual-branch conditioner and adaptive modulation to guide a large

diffusion transformer (Flux. 1) with minimal overhead, achieving robust, high-fidelity restoration without relying on text prompts or MLLM captions. 🔧 Installation Inference Prepare models in 2 steps, then run a single command. 1) Login to Hugging Face (required for gated FLUX. 2) Download required weights to fixed paths and export env vars Run inference (uses fixed relative paths): ℹ️ LucidFlux builds on Flux-based generative priors. Restored images can differ from the low-quality input because the model removes degradations and

hallucinates realistic details by design. Visual discrepancies are expected and indicate the generative nature of the method. You can also obtain results of LucidFlux on RealSR and RealLQ250 from Hugging Face: **LucidFlux**. Data and Filtering Pipeline We have released the metadata for the clean images used in LucidFlux at LucidFlux-Training-Data, and the filtering pipeline in . Images can be downloaded from the provided , and filtering scores can be computed with: The LSDIR dataset can be downloaded from LSDIR, and

degradation synthesis follows the pipeline from Real-ESRGAN. If more data is needed, please also refer to our MultiAspect-4K-1M dataset. The training data can be prepared in the same structure as the example under , where each sample is organized as paired images in and . In our current setup, the training run takes roughly 50 GB VRAM with , and about 43 GB with . Once the data is ready, you can start training with: 🚀 Updates For the purpose of fostering

research and the open-source community, we plan to open-source the entire project, encompassing training, inference, weights, etc. Thank you for your patience and support! [x] Release the training data and filtering pipeline. 🙏 Acknowledgments This code is based on FLUX. Some code are brought from DreamClear, x-flux. We thank the authors for their awesome work. 🏛️ Thanks to our affiliated institutions for their support. 🤝 Special thanks to the open-source community for inspiration.