Paper page - Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance

Source: https://huggingface.co/papers/2606.19195

Abstract

A lightweight image inpainting framework achieves high-fidelity results with significantly reduced parameters and inference time through novel local-global interaction blocks and adaptive distillation strategies.

While 10B-level industrial foundation models have pushed the boundaries of image inpainting, their prohibitive computational costs severely hinder practical deployment. Constructing a highly optimized task-specific specialist offers a promising solution; however, extreme structural compression inevitably triggers a severerepresentation bottleneck. To conquer this, we propose Moebius, a highly efficient lightweight inpainting framework. We systematically reconstruct thediffusion backboneby introducing theLocal-λ Mix Interaction(LλMI) block. Comprising Local-λ and Interactive-λ modules, it elegantly summarizesspatial contextsandglobal semantic priorsintofixed-size linear matrices, preserving complex latent interactions while drastically shedding parameters. Furthermore, to unlock the full representational capacity of this highly compact architecture, we synergistically pair it with anadaptive multi-granularity distillationstrategy. Operating strictly within thelatent spaceto avoid expensive pixel-space decoding, this strategy dynamically balances multiplegradient-based lossesto achieve high-fidelity alignment. Extensive experiments across natural and portrait benchmarks demonstrate that this optimal synergy enables Moebius to rival or even surpass the generation quality of the 10B-level industrial generalist FLUX.1-Fill-Dev. Remarkably, Moebius achieves this using less than 2\% of the parameters (0.22B vs. 11.9B) while delivering a >15times acceleration in total inference time, setting a new efficiency standard forhigh-fidelity inpainting. Project page at https://hustvl.github.io/Moebius.

View arXiv pageView PDFProject pageGitHub66Add to collection

Get this paper in your agent:

hf papers read 2606\.19195

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.19195 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.19195 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.19195 in a Space README.md to link it from this page.

Collections including this paper2