donahowe/AutoStudio
AutoStudio is a training-free multi-agent framework that uses LLM-based agents to orchestrate consistent multi-turn interactive image generation with Stable Diffusion.
Not currently ranked — collecting fresh signals.
star history
AutoStudio enables users to interact with a system over multiple turns to generate coherent image sequences while maintaining subject consistency. The framework employs three LLM-based agents (subject manager, layout generator, supervisor) and a Stable Diffusion-based drawer agent. It introduces a Parallel-UNet architecture with dual cross-attention modules for subject-aware feature exploitation and a subject-initialized generation method to preserve small subjects across image generations.
Frequently asked
- What is donahowe/AutoStudio?
- AutoStudio is a training-free multi-agent framework that uses LLM-based agents to orchestrate consistent multi-turn interactive image generation with Stable Diffusion.
- Is AutoStudio open source?
- Yes — donahowe/AutoStudio is an open-source project tracked on heatdrop.
- What language is AutoStudio written in?
- donahowe/AutoStudio is primarily written in Jupyter Notebook.
- How popular is AutoStudio?
- donahowe/AutoStudio has 449 stars on GitHub.
- Where can I find AutoStudio?
- donahowe/AutoStudio is on GitHub at https://github.com/donahowe/AutoStudio.