Show HN: An AI-first visual editor using GPT-4o's GPT-image-1 model
img.lyHey HN! We’ve been experimenting with integrating multimodal models directly into creative workflows, and ended up building an AI-first image editor using OpenAI’s new `gpt-image-1` (from GPT-4o) inside our SDK.
Instead of prompting in ChatGPT and pasting outputs into a design tool, this lets you generate, edit, and remix images all in one canvas. This allows for really interesting new workflows, like quickly mixing multiple images, or creating visual prompts by using annotations and reference on the canvas.
Some key details:
- Built with our plugin system in CE.SDK (CreativeEditor SDK) - Works with any multi modal model/API, not just OpenAI - You can bring your own API key, or extend it however you like - Focused on real-world workflows, not just concept generation
We put together a live demo and wrote a bit more on why we built it.
Would love feedback or to hear how others are thinking about AI in creative tools. If you have questions or ideas, let us know! I'm curious about the architecture behind this integration — how does the system handle communication between GPT-4o and the visual editor in real time? Are there any challenges in syncing AI-generated content with editable UI elements? Really interesting! I noticed it takes quite a long to generate images, what’s your take on managing latency? Hey, yes, that's somewhat unavoidable given that is simply takes OpenAI that long to generate images. I think that's to be expected by most users. We tried to be as friendly as possible to parallelization so users aren't blocked while working in the editor. I would assume openAI bring in the delay? Looks very cool, any plans to open source parts of this or is it all part of your SDK? Cool! How does this compare to what tools like Canva or Adobe offer for automation? Well, it's fundamentally inspired by what these companies are trying to do. However, we are an SDK so we want to offer a toolset for other apps to build similar experiences. impressive...can't wait for OpenAI to make their image gen faster tough