GenRouter：面向智能體圖像生成的統一工作流路由

摘要

文字轉圖像（T2I）生成模型的快速演進已有效解決了原始像素合成這項基礎挑戰，使社群的研究焦點轉向滿足日益複雜的使用者需求。雖然近期的代理式影像生成工作流程透過外部知識檢索與迭代推理等先進能力增強了靜態推論，但它們大多在彼此孤立的環境中運作，採用固定的「一體適用」拓撲結構。這不可避免地導致嚴重的計算不匹配，即簡單查詢被迫通過計算密集的管線處理。為彌補此一差距，我們提出 GenRouter，這是首個統一代理式影像生成的工作流程路由框架。我們首先建構 GenCanvas，將多樣化的代理式管線標準化為一組通用的基礎原語與可執行模板。在此統一空間中運作的 GenRouter 透過（i）需求剖析、（ii）經驗匹配及（iii）Pareto 過濾，自適應地將異質提示路由至其最佳工作流程。跨多樣化基準的廣泛實驗表明，與重量級靜態管線相比，GenRouter 在實現卓越視覺對齊的同時，將執行成本降低超過 95%，延遲降低 65%。此外，該系統透過累積經驗持續自我進化，實現穩健的零樣本泛化能力，在提升效能的同時將計算開銷減半。

English

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 recent agentic image generation workflows 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 present GenRouter, the first unified workflow routing framework for agentic image generation. We first formulate GenCanvas, standardizing diverse agentic pipelines into a universal set of foundational primitives and executable templates. Operating over this unified space, GenRouter adaptively routes heterogeneous prompts to their optimal workflows via (i) demand profiling, (ii) experience matching, and (iii) Pareto filtering. Extensive experiments across diverse benchmarks demonstrate that GenRouter achieves 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 robust zero-shot generalization that boosts performance and halves computational overhead.