# ComfyGI: Automatic Improvement of Image Generation Workflows

arXiv: https://arxiv.org/abs/2411.14193 · HTML: https://mlsanity.com/paper/2411.14193/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 44 references that are in the map.
- First posted on arXiv 21 Nov 2024.

## Builds on (most cited first, 6 of 6)

- [IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2308.06721/) (2023)
- [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://mlsanity.com/paper/2304.06767/) (2023)
- [Aligning Text-to-Image Models using Human Feedback](https://mlsanity.com/paper/2302.12192/) (2023)
- [DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models](https://mlsanity.com/paper/2210.14896/) (2022)
- [On Discrete Prompt Optimization for Diffusion Models](https://mlsanity.com/paper/2407.01606/) (2024)
- [Crafting Generative Art through Genetic Improvement: Managing Creative Outputs in Diverse Fitness Landscapes](https://mlsanity.com/paper/2407.20095/) (2024)

## Built on by (most cited first, 5 of 5)

- [ComfyUI-R1: Exploring Reasoning Models for Workflow Generation](https://mlsanity.com/paper/2506.09790/) (2025)
- [ComfyUI-Copilot: An Intelligent Assistant for Automated Workflow Development](https://mlsanity.com/paper/2506.05010/) (2025)
- [COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows](https://mlsanity.com/paper/2607.01709/) (2026)
- [Knowledge-Centric Agents for Workflow Generation in ComfyUI](https://mlsanity.com/paper/2607.15845/) (2026)
- [Policy Optimized Text-to-Image Pipeline Design](https://mlsanity.com/paper/2505.21478/) (2025)

## Abstract (arXiv)

Automatic image generation is no longer just of interest to researchers, but also to practitioners. However, current models are sensitive to the settings used and automatic optimization methods often require human involvement. To bridge this gap, we introduce ComfyGI, a novel approach to automatically improve workflows for image generation without the need for human intervention driven by techniques from genetic improvement. This enables image generation with significantly higher quality in terms of the alignment with the given description and the perceived aesthetics. On the performance side, we find that overall, the images generated with an optimized workflow are about 50% better compared to the initial workflow in terms of the median ImageReward score. These already good results are even surpassed in our human evaluation, as the participants preferred the images improved by ComfyGI in around 90% of the cases.