Paper page - GenRouter: Unified Workflow Routing for Agentic Image Generation

Source: https://huggingface.co/papers/2608.16721

Abstract

GenRouter is a unified routing framework that adaptively directs prompts to optimal agentic image-generation workflows, cutting costs and latency while improving visual alignment and enabling continuous self-evolution.

The rapid evolution of text-to-image (T2I) generation models has effectively solved the foundational challenge of raw pixel synthesis, shifting the community’s focus toward fulfilling increasingly intricate user requests. While recentagentic image generationworkflows enhance static inference with advanced capabilities like external knowledge retrieval and iterative reasoning, they mostly operate in isolated silos with fixed ``one-size-fits-all“ topologies. This inevitably leads to severe compute-mismatch, where simple queries are forced through computationally heavy pipelines. To bridge this gap, we presentGenRouter, the first unifiedworkflow routingframework foragentic image generation. We first formulateGenCanvas, standardizing diverse agentic pipelines into a universal set of foundational primitives and executable templates. Operating over this unified space,GenRouteradaptively routes heterogeneous prompts to their optimal workflows via (i)demand profiling, (ii)experience matching, and (iii)Pareto filtering. Extensive experiments across diverse benchmarks demonstrate thatGenRouterachieves superior visual alignment while reducing execution costs by over 95% and latency by 65% compared to heavyweight static pipelines. Furthermore, the system continuously self-evolves via accumulated experience, enabling robustzero-shot generalizationthat boosts performance and halves computational overhead.

View arXiv pageView PDFGitHub11Add to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.16721 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.16721 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.16721 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.