Announcements

This month's update brings on-device removal powered by a downloadable AI model, the Remove Tool's ability to detect signs, poles, and other distractions beyond people, Live Filter Layers for non-destructive filtering directly in the layer stack, and the ability to send documents straight to Firefly Boards. A broad set of generative AI, interface, and file handling bugs have also been resolved. Feature Updates & Additions On-Device Remove ToolYou can now download an AI model (approximately 4 GB) to run generative removals entirely on your device, with no internet connection required. Processing is faster, works offline, and is well suited for enterprise or restricted-network environments. On NVIDIA-equipped systems, the model now correctly uses GPU acceleration for optimal performance.General Distractors DetectionThe Remove Tool can now automatically detect and remove signs, poles, trash, and other unwanted objects, not just people and wires. One-click removal handles a broader ran

We are excited to announce the beta launch of an on-device generative AI model for the Remove Tool!You can now download the Generative AI model onto your personal device and work off-line and process locally -  ideal when internet connections are unreliable or when cloud processing isn’t preferred. (It’s kind of a big deal!) How to:Select the Remove Tool. In the Options bar, open the Mode drop‑down (Auto / GenAI On / GenAI Off). In the Generative AI Processing section, choose Device.If the Device option is dimmed and unavailable, your system does not meet the minimum requirements to run the model. When prompted, click Download to start the download (approximately 5 GB). Once the download completes, the on‑device model will be used automatically. Switch between Cloud and Device processing at any time from the Mode drop‑down.Choose the Remove Tool, then visit the Mode drop-down in the Options Bar to find the Generative AI processing options.System RequirementsThis is a GPU‑intensive feat

We are excited to announce the latest updates to the Actions panel which are now in Photoshop beta!As part of our continuous efforts to improve the Actions panel, we have introduced some changes to help streamline your editing workflows. 🆕 Introducing the Essentials tabWe're excited to unveil the rebranded "Essentials" tab (formerly known as the "Suggestions" tab). The new Essentials tab provides a new default set of categorized actions, created to help simplify common editing workflows from edit to export. Check it out!Go to Window > Actions. Browse through the Essentials set of actions and select the ones that fit your workflow.  Essentials categories include: Basic Adjustments, Subject & Background, Creative Effects, Guides, Resize, and Export.  Please try out the new Essential actions and share your feedback with us! Thank you! 

This month's update includes a redesigned Generative Fill interface with reference image support and an updated model picker in the contextual task bar. Rotate Object received visual and interaction refinements, and a broad set of stability and workflow issues have been resolved across Generative AI, Adjustments, Actions, and Type. Feature Updates & Additions Generative Fill: Redesigned InterfaceThe Generative Fill dialog has been redesigned. Reference images are now accessible directly in the interface, with no need to switch to a separate panel. The model picker now appears in the contextual task bar alongside the prompt. A curated set of models replaces the previous long scrolling list, making it faster to switch between Firefly, Gemini, and other available options. Your generative AI credit balance is now visible in the contextual task bar while you work. Firefly Image 3 is no longer available in the Generative Fill model picker. The new default model is Firefly Fill & Expa

Quickly clean-up unwanted objects that pull attention away from your main subject.General Distraction Removal (beta) in the Remove Tool automatically finds and removes common distractions—like poles, vehicles, and visual clutter—with just a few clicks.How to try it in Photoshop (beta) for desktop:1. Select the Remove Tool from the toolbar.2. In the Options bar, click Find Distractions.3. Click Find next to General Distractions.A color‑coded list of detected distractions appears, and each distraction is highlighted on the canvas with a matching overlay. You can change the color of the distraction overlay by clicking on the color swatch in the panel.   Select the Remove Tool, then choose Find General Distractions from the Options Bar to see color‑coded results and editable overlays  4. Use the checkboxes to turn targeting on or off for each distraction category.You can also refine the results by brushing directly on the canvas using the + / − controls in the Remove Tool options bar.   

Firefly Image 5 enables a new instruction‑based editing workflow in Generative Fill and it's currently in Photoshop (Beta) for desktopFirefly Image 5 is Adobe's most advanced image generation and editing model yet; you can edit images more intuitively and precisely with just a prompt. Simply describe your edits with natural language instructions to transform the entire image—such as changing time of day or modifying a subject’s pose - making the precise changes you want with high quality results, while keeping the rest of your pixels and the details of your original image the same. How to try it:1. Open an image and select the entire canvas (Select > Select All…).(While smaller selections are supported, this beta performs best when the full image is selected.)2. Click Generative Fill in the Contextual Task Bar.3.Choose Firefly Image 5 (Preview) from the model picker in the Contextual Task Bar.4. Enter a clear, instructional prompt (for example: “change the time of day to sunset”).5.

We’re excited to share that Firefly Boards is now more closely integrated with Photoshop workflows, enabling seamless ideation, organization, and round-trip editing. What is Firefly Boards?Firefly Boards is an AI-powered ideation workspace that helps Photoshop users capture inspiration, explore multiple visual directions using the latest AI models, and organize ideas on an infinite canvas. All linked assets stay synced, so updates reflect directly on the Boards canvas.Try it out:Home > Your Files > 3-dot menu > Open in BoardsShare icon > Send to Firefly Boards (PSDC files)File > Export > Send to Firefly Boards (PSDC files)Round-trip workflow from Firefly Boards → Photoshop (Adobe Firefly Boards) We’d love for you to try Firefly Boards for ideation and share your feedback!

Rotate Object lets you quickly transform 2D objects into rotatable 3D images, allowing rotation (spin, tilt, dolly) from any viewpoint to match perspective and depth. (Version: 27.5.0 20260311.m.3453)HOW TO TEST:1. Open or place an image 2. Select a pixel layer that contains an object. PRO TIP: You may want to duplicate the layer → as the layer will be converted to a Rotate Object LayerCmd + T to enter transform mode or Edit > Rotate Object Click the Rotate object button in the context bar - note this feature is 20 Generative Credits, the first 3 tries do not deduct generative credits.3. After the 3D conversion is complete, the initial rotated object will be a low-res PREVIEW. The layer was converted to a Rotate object layer. DIFFERENT WAYS TO ROTATE AN OBJECT:  Contextual Task Bar: click to use the Rotate and Tilt sliders. Click and drag the blue on-canvas controls around the object. Right-click on the object and drag (on-canvas for unconstrained rotation). Properties panel:

Generative Fill 3rd party partner models such as Gemini 2.5, 3.0 (with Nano Banana), FLUX Kontext Pro, and FLUX.2 Pro can use reference images as inputs that it can "read" and apply according to your prompt instructions.The party partner model scans all uploaded images for identity, style, and structure, then matches your prompt instructions to specific parts of those images. When complete, the 3rd party partner model generates a new image that merges the subject and style from the reference images.  HOW TO TEST Open an image > Select All  Select > Generative Fill on Context Bar or Edit > Generative Fill Select either Nano 2.5, 3.0, FLUX Kontext Pro, FLUX.2 Pro ENTER A PROMPT in the text field (Example prompt: Replace the model’s clothing with the 6 reference images) Select the Reference Image icon (icon to the right of the model picker dropdown) Add up to 8 reference images if using Gemini 3 > Generate HOW TO PROVIDE FEEDBACK:1. Click on the Beaker Icon in the upper right

The AI-powered Reflection Removal feature lets you easily extract and remove reflections that impact the majority of your scene. The Reflections Removal command eliminates large and prominent window reflections in just a few steps, allowing your image's subject to stand out clearly. While AI-powered, it is not generative AI and does not use credits.How to:Open a file with reflections that dominate the image. Edit > Reflections Removal…The Reflection Removal Dialog appears Choose the quality of the results from the pop up.    Turning on "Create separate reflection layer" will give you both the subject "Reflection Removal" layer and a layer with the extracted reflection. Click [OK] The  "Reflection Removal" layer and, if checked ON, the Reflection layer appear in the Layers panel. Adjust the Layer Opacity of the subject "Reflection Removal" layer to bring back subtle reflections if desired.  Adjust the Opacity slider in the layers panel to bring back subtle reflections for a natural