ComfySearch: Autonomous Exploration and Reasoning for ComfyUI Workflows
This addresses the challenge for users of ComfyUI in creating functional and high-quality AI-generated content workflows, but it is incremental as it builds on existing agentic and validation methods.
The paper tackled the problem of low pass rates and limited quality in ComfyUI workflows due to many components and strict graph constraints, resulting in ComfySearch, an agentic framework that outperforms existing methods with higher executability and solution rates.
AI-generated content has progressed from monolithic models to modular workflows, especially on platforms like ComfyUI, allowing users to customize complex creative pipelines. However, the large number of components in ComfyUI and the difficulty of maintaining long-horizon structural consistency under strict graph constraints frequently lead to low pass rates and workflows of limited quality. To tackle these limitations, we present ComfySearch, an agentic framework that can effectively explore the component space and generate functional ComfyUI pipelines via validation-guided workflow construction. Experiments demonstrate that ComfySearch substantially outperforms existing methods on complex and creative tasks, achieving higher executability (pass) rates, higher solution rates, and stronger generalization.