[RFC]: OpenAI Image Edit API Interface for ComfyUI

[lishunyang12]

Motivation.

While ComfyUI supports local inpainting and OpenAI text-to-image, it lacks a native or well-standardized way to leverage OpenAI's cloud-based editing. Users currently have to write custom scripts or use fragmented third-party nodes. Adding this enables:

Proposed Change.

Propose the implementation of a standardized node and backend interface to support the OpenAI v1/images/edit endpoint. This will allow users to perform prompt-based image editing (inpainting and outpainting) using DALL-E 2 or GPT-Image models directly within ComfyUI workflows.

High-quality inpainting for users without powerful local GPUs.

Consistency with the existing OpenAI T2I implementation.

Hybrid Workflows: Using local Stable Diffusion for base generation and DALL-E/GPT-Image for final "magic" edits.

Feedback Period.

No response

CC List.

@david6666666 @dougbtv

Any Other Things.

No response

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

[david6666666]

This looks great!

[fhfuih]

Do we currently support /images/edit endpoint? Or is this RFC exactly talking about the potential implementation of this endpoint? I notice that our current online image editing routine still uses the chat completion endpoint

[david6666666]

closed by #1101