Polaris: Scaling Up Instruction-Guided Image Generation Towards Millions of Personalized Style Needs

2026-06-01

AI Summary

Existing image generation models struggle to efficiently and cost-effectively fulfill users’ highly personalized and diverse stylistic demands. This work proposes an intelligent retrieval framework that, for the first time, systematically constructs and leverages a heterogeneous ecosystem of over 6,500 checkpoints and 75,000 adapters from pretrained models. By building large-scale indexes, the framework enables instruction-driven model retrieval, dynamic multi-module composition, and alignment between instructions and models. Without requiring any additional training or fine-tuning, the method adaptively assembles models on demand to achieve highly controllable, consistent, and scalable personalized image generation and editing.

Abstract

Users increasingly expect image generation models to quickly adapt to highly diverse and personalized requirements, such as producing images with distinctive styles or characteristics. Traditional approaches rely on fine-tuning, which is costly and difficult to scale. To cope with these limitations, the community has accumulated a growing library of fine-tuned modules and adapters, where each component targets specific generation needs and collectively serves as a foundation for handling new demands. This naturally raises a question: instead of repeatedly training new models, can we systematically exploit this expanding ecosystem to better fulfill user instructions? To this end, we present Polaris, an intelligent retrieval framework that automatically selects and integrates suitable models from the model library based on a user's instructions. The key insight is that harnessing such a massive and heterogeneous pool requires not only finding the most relevant modules among thousands of candidates, but also aligning them effectively for instruction-driven generation and editing. Polaris addresses this challenge by indexing over 6,500 checkpoints and 75,000 adapters, and retrieving the most relevant components given a user's input and instruction. In doing so, it delivers scalable, controllable, and well-aligned generation -- without any additional training.

Problem

• personalized image generation
• instruction-guided generation
• model library
• scalability
• style adaptation

Innovation

• instruction-guided generation
• model retrieval
• personalized image generation
• adapter composition
• scalable AI