VisionCreator: A Native Visual-Generation Agentic Model with Understanding, Thinking, Planning and Creation
This work addresses the problem of autonomous creative planning in visual content creation for researchers and developers in the field of artificial intelligence and computer vision, providing a foundation for future research in visual-generation agentic systems.
The authors tackled the challenge of visual content creation by proposing VisionCreator, a model that unifies understanding, thinking, planning, and creation capabilities, achieving superior performance over larger models. The VisionCreator-8B/32B models demonstrated strong results, outperforming larger closed-source models across multiple evaluation dimensions.
Visual content creation tasks demand a nuanced understanding of design conventions and creative workflows-capabilities challenging for general models, while workflow-based agents lack specialized knowledge for autonomous creative planning. To overcome these challenges, we propose VisionCreator, a native visual-generation agentic model that unifies Understanding, Thinking, Planning, and Creation (UTPC) capabilities within an end-to-end learnable framework. Our work introduces four key contributions: (i) VisGenData-4k and its construction methodology using metacognition-based VisionAgent to generate high-quality creation trajectories with explicit UTPC structures; (ii) The VisionCreator agentic model, optimized through Progressive Specialization Training (PST) and Virtual Reinforcement Learning (VRL) within a high-fidelity simulated environment, enabling stable and efficient acquisition of UTPC capabilities for complex creation tasks; (iii) VisGenBench, a comprehensive benchmark featuring 1.2k test samples across diverse scenarios for standardized evaluation of multi-step visual creation capabilities; (iv) Remarkably, our VisionCreator-8B/32B models demonstrate superior performance over larger closed-source models across multiple evaluation dimensions. Overall, this work provides a foundation for future research in visual-generation agentic systems.