product update

Adobe Tests AI Photo Editing in Project Indigo App Using Google's Nano Banana Model

TL;DR

Adobe is testing AI-powered photo editing features in its experimental Project Indigo camera app for iPhone. The AI Playground, available to a small percentage of users for a limited time, includes object removal, style transfers, photo critiques, and custom prompt-based editing powered by Google's Nano Banana model.

2 min read
0

Adobe Tests AI Photo Editing in Project Indigo App Using Google's Nano Banana Model

Adobe is rolling out experimental AI photo editing tools to select users of Project Indigo, its free iPhone camera app that launched in 2025. The new "AI Playground" feature set uses Google's Nano Banana model as its underlying engine.

Limited Test Parameters

The AI Playground will be available to "a few percent" of Project Indigo users for "a few weeks," according to Adobe. The company is offering the features free with no sign-in required during this experimental phase. Adobe states its goal is to understand how users interact with generative AI editing tools. If successful, the company plans to eventually offer a paid version.

Adobe says it will only collect button press analytics during the test. The company claims it "won't look at your prompts, or images, or upload them to [its] servers, or use them to train AI models." Data collection is anonymous due to the lack of required sign-in.

Four AI Editing Categories

The AI Playground includes:

Object Editing: Removes distractions from photos and can defocus backgrounds to simulate shallow depth of field. Users can toggle removal of predefined distractor categories or add custom ones via text or voice input.

Styles: Applies artistic techniques to photos, re-rendering images to mimic traditional art styles. Adobe says these represent techniques that work best on everyday snapshots.

Photo Guidance: Provides AI-generated critiques and suggestions for both reshoot and editing improvements, designed to help users learn photography.

Custom Edit: Accepts arbitrary text prompts for generative editing. Adobe specifically mentions the capability to convert drawings, paintings, or sculptures into photographic representations.

Technical Implementation

Adobe is using Google's Nano Banana model, which outputs Standard Dynamic Range (SDR) images only. Since Project Indigo supports High Dynamic Range (HDR) capture, Adobe built a custom machine learning model to upconvert Nano Banana's SDR output back to HDR. The company acknowledges this process "is not perfect and relies heavily on the output Nano Banana produces."

The AI features only work on images captured within Project Indigo. For imported photos, the AI Playground button remains disabled.

App Background

Project Indigo launched over a year ago as an experimental iPhone camera app with full manual controls. In April 2026, Adobe expanded it to select iPads and the iPhone 17e, adding grid view, photo filtering, and quality improvements.

What This Means

Adobe is testing consumer appetite for AI editing in a prosumer camera app while avoiding the privacy concerns that typically accompany cloud-based generative AI. The use of Google's Nano Banana model—presumably an on-device model given the no-upload claims—represents a shift toward local AI processing. The custom HDR upconversion layer Adobe built suggests current on-device models still lag behind cloud alternatives in image quality. The time-limited, small-scale test indicates Adobe is cautious about committing resources before validating user interest.

Related Articles

product update

Google releases free AI Edge Foresight, a Mac note-taking app that transcribes meetings fully offline

Google has released AI Edge Foresight, a free experimental macOS app that transcribes meetings and audio files entirely on-device. It is optimized for Apple Silicon Macs and uses Google's EmbeddingGemma 2 model, according to The Verge. Google says audio, files, and notes never leave the user's computer.

product update

Google rolls out Gemini 4 Argon to Antigravity for AI Pro and Ultra subscribers with separate weekly quota

Google is rolling out Gemini 4 Argon, its latest flagship model, to the Antigravity agentic development platform for AI Pro and AI Ultra subscribers. Argon has a separate weekly-only quota on top of the existing Gemini 3.8 Flash quota. Google also published internal case studies, which are company-reported and not independently verified.

product update

Liquid AI Publishes GGUF Builds of LFM2.5-24B-A2B, Including a 13.6 GB Quantization-Aware Q4_0 Checkpoint

Liquid AI has published GGUF builds of LFM2.5-24B-A2B, a 24B-parameter model on the lfm2moe architecture, in sizes from 13.6 GB (Q4_0) to 48 GB (BF16). The Q4_0 file is a Quantization-Aware Distillation checkpoint trained to recover accuracy lost to 4-bit quantization.

product update

Google builds Antigravity into Gemini app for macOS, limited to AI Ultra subscribers

Google has integrated Antigravity, its agent-first development platform, directly into the Gemini app for macOS as part of the Gemini 4 Argon rollout. Access requires an AI Ultra subscription, while the new model itself requires AI Pro or above. The update also renames and reorganizes several app features, including Gemini Spark, Skills and Connectors.

Comments

Loading...