Apple's New AI Dataset Aims to Improve Photo Editing Models - MacRumorsOpen MenuShow RoundupsShow Forums menuVisit ForumsOpen Sidebar
Skip to Content

Apple's New AI Dataset Aims to Improve Photo Editing Models

Apple researchers have released Pico-Banana-400K, a comprehensive dataset of 400,000 curated images that's been specifically designed to improve how AI systems edit photos based on text prompts.

apple intelligence black
The massive dataset aims to address what Apple describes as a gap in current AI image editing training. While systems like GPT-4o can make impressive edits, the researchers say progress has been limited by inadequate training data built from real photographs. Apple's new dataset aims to improve the situation.

Pico-Banana-400K features images organized into 35 different edit types across eight categories, from basic adjustments like color changes to complex transformations such as converting people into Pixar-style characters or LEGO figures. Each image went through Apple's AI-powered quality control system, with Google's Gemini-2.5-Pro being used to evaluate the results based on instruction compliance and technical quality.

The dataset also includes three specialized subsets: 258,000 single-edit examples for basic training, 56,000 preference pairs comparing successful and failed edits, and 72,000 multi-turn sequences showing how images evolve through multiple consecutive edits.

Apple built the dataset using Google's Gemini-2.5-Flash-Image (aka Nano-Banana) editing model, which was released just a few months ago. However, Apple's research revealed its limitations. While global style changes succeeded 93% of the time, precise tasks like relocating objects or editing text seriously struggled, with success rates below 60%.

apple image editing ai dataset pico banana
Despite the limitations, researchers say their aim with Pico-Banana-400K is to establish "a robust foundation for training and benchmarking the next generation of text-guided image editing models." The complete dataset is freely available for non-commercial research use on GitHub, so developers can use it to train more capable image editing AI.

Top Rated Comments

JSRinUK Avatar
22 weeks ago
Ah, with Apple’s involvement I enthusiastically await the “Photo Tools” equivalent of this popping up whenever I want to use it…



Attachment Image
Score: 11 Votes (Like | Disagree)
mjs916 Avatar
22 weeks ago
I realize this isn’t the point but… Can Apple’s own AI tools do what Gemini can? I’ve been unimpressed by Image Playground.
Score: 11 Votes (Like | Disagree)
turbineseaplane Avatar
22 weeks ago
Ok ... sure 🤷‍♂️

I'm very much in the "show me, don't tell me" camp with Apple and any of their AI / Siri efforts.
Score: 11 Votes (Like | Disagree)
rp2011 Avatar
22 weeks ago
Anyone who has used any of the AI tools knows none of them are ready for Prime Time. None of them. That is why I always say it's just echo-chamber nonsense to say Apple or anyone is behind in this nacent quickly evolving open source media. That ALL of them have been RUSHED to market without adequate testing goes without saying.

There is immense room for improvement with just the low hanging fruit of better training data. And here we see Apple showing they certainly know that.
Score: 10 Votes (Like | Disagree)
everlast3434 Avatar
22 weeks ago
Converting people into Legos. Seriously lol. Do people want this stuff?
Score: 8 Votes (Like | Disagree)
Mac Fly (film) Avatar
22 weeks ago

Anyone who has used any of the AI tools knows none of them are ready for Prime Time. None of them. That is why I always say it's just echo-chamber nonsense to say Apple or anyone is behind in this nacent quickly evolving open source media. That ALL of them have been RUSHED to market without adequate testing goes without saying.

There is immense room for improvement with just the low hanging fruit of better training data. And here we see Apple showing they certainly know that.
100%

All of the LLMs are confidently wrong the whole time.
Score: 6 Votes (Like | Disagree)