Adobe Integrates Google's Nano Banana AI Model into Photoshop and Firefly

Photoshop Is Adding Google's Nano Banana to Its Most Popular AI Editing Tool

Photoshop is getting its first-ever non-Adobe AI models, starting with Gemini 2.5 Flash Image (better known by its nickname, nano banana) and Flux.1 Kontext Pro. These new, third-party models give creators who use the generative fill tool more options for AI media generation without having to leave their current project. They're only available in the Photoshop beta app starting today, but many beta features eventually make their way into the regular version of Photoshop. Anyone who has used AI knows it makes mistakes, called hallucinations. By using AI image generation in the Photoshop app, the tools necessary for fixing those mistakes are right at hand. "You have the full power of Photoshop around you. So as you use generative fill with these [new] choices, you can also continue to refine your results using the powerful engine that is Photoshop: layers, masks, selections," Deepa Subramaniam, vice president of creative professionals marketing for Adobe, said in an interview. "So it's really about how all of this comes together within Photoshop." This is also good news for Adobe subscribers who want to play around with the AI darling du jour, Google's nano banana. This new image editing model has drawn plenty of attention and excitement, and it's already available in other Adobe programs, including Express and Firefly, as well as Google's AI Studio. Standard subscribers (those on the Creative Cloud standard, photography plan and Photoshop desktop single app plan) will get 100 lifetime generation credits for each model, while Creative Cloud Pro subscribers get 500 credits per day for each model. But from now until Oct. 28, you can use the models without deducting credits. Right now, you can only select outside models when using generative fill, though Subramaniam said there are plans to bring outside models to other AI tools in the future. Generative fill lets users select specific areas and use text prompts for AI to "fill" the area with AI-generated imagery. Outside AI models fall under Adobe's AI policy, which promises not to train on your content. The company does warn that third-party material may not be OK for commercial use, unlike content made with its in-house Firefly models. When you go to select those regions, you can swap out Adobe Firefly's image model for Gemini or Flux. You can confine changes to select areas with your select tool, something that's hard to do with other AI image generation tools. You can also take the opposite approach by selecting the entire project and using text prompts to have Gemini or Flux reimagine it, with your existing assets serving as reference images. Photoshop has added many AI tools over the past few years, to the chagrin of some users. I've tested those features extensively as CNET's AI image and video expert. Adobe has long offered image generation through its Firefly AI models, integrated into Photoshop and other programs. But Photoshop's AI updates have mostly focused on editing rather than wholesale creation. Generative remove is an AI heavy-duty eraser, for example. You can check out this guide to using Photoshop's AI tools for more tips. Don't miss any of our unbiased tech content and lab-based reviews. Add CNET as a preferred Google source. Adobe has been expanding its roster of third-party AI models over the past few months. Here are all the models you can use in Adobe programs.

You can now use Google's viral Nano Banana AI directly in Photoshop

* Google's new Mixboard offers AI-fueled visualization and image remixing * Users can create and edit idea boards using text prompts * Mixboard is available as an experimental beta in the U.S. Google's popular new Gemini 2.5 Flash Image model, better known as Nano Banana, has a new home in Adobe Photoshop. Those with access to Photoshop's beta app can produce images with Nano Banana, best known for making viral 3D figurines from photographs, directly within the Generative Fill tool, alongside Adobe's own Firefly model. Nano Banana is joining the platform along with a new contextual model from Black Forest Labs, the developers of the AI image-maker used by Grok on X, called FLUX.1 Kontext [pro]. It's a notable expansion for Adobe's AI strategy, which has largely relied on Adobe-built models until now. But now, Photoshop users can generate a dreamy, stylized background using Nano Banana, refine it with realistic shadows via FLUX.1, and polish the whole thing using Firefly, all within the Generative Fill feature. Adobe seems to want Photoshop to serve as a hub for multiple AI models to compete for your creative attention. Adobe still encourages users to consider the Firefly model as the go-to for commercial-safe results. Firefly is built for higher-end images in terms of quality, consistency, and reliance on a database that Adobe has permission to use for training its models. Nano Banana and FLUX have their own strengths, of course. Google's contribution is known for its stylistic flourishes and eye-catching visuals, which are based on text, while Black Forest Labs pitches its model as ideal for more realistic and coherent images. Now that they're swappable inside the same prompt window, users will be able to test those strengths out very quickly. For instance, if you have an idea for a fantasy story illustration, Nano Banana could add some magical effects, while FLUX realistically lights a cave setting, and Firefly cleans it all up. AI image imagination Adobe opening up to third-party providers will likely help the company as it battles to keep Photoshop atop the list of high-end visual platforms. Making it easy to switch among the models will streamline a lot of projects, both professional and personal. Speeding up the sometimes tedious process of editing images is going to appeal to plenty of creators. For instance, a designer making social content can generate stylized backgrounds in seconds and tweak them without toggling between tabs. An inventor could run the entire product imagery and marketing launch campaign entirely within Photoshop, and even someone making a funny greeting card image will find the process a lot easier. You still need taste and direction. However, the burden of drafting and editing might be significantly lighter. If you want to try mixing Photoshop and Nano Banana, you'll need to opt into the Photoshop beta via Creative Cloud. Adobe says more improvements are coming as they optimize the back-end systems and introduce finer controls over each model's output.

Photoshop Has Added Google's Viral Nano Banana AI Model to Generative Fill

Adobe has added two popular AI image models to Photoshop Beta, including Google's viral Gemini 2.5 Flash Image, otherwise known as Nano Banana. Adobe says that Generative Fill has become one of Photoshop's most popular tools, and it is expanding its capabilities by adding Nano Banana and Black Forest Labs' FLUX.1. After clicking on Generative Fill, the dialog box will appear, and there is an option to select either Google's Nano Banana, FLUX, or Adobe's Firefly model. "By combining the power of Adobe Photoshop with the best AI models, you have access to the most helpful creative AI tools where they're most helpful to you as you express your vision to the world," Adobe says. "What makes this integration into Photoshop especially powerful is how it combines choice in models with Photoshop's professional-grade tools, giving you the ability to generate content, then immediately refine it with Photoshop's precision and control. It's a seamless connection that helps you move from idea to impact even faster without breaking your creative flow." Adobe says that Nano Banana is best for "stylized elements, graphic details, and imaginative scene additions." While FLUX.1 is designed for "contextual accuracy, perspective, and environmental harmony." Meanwhile, Adobe says its own Firefly image models are "commercially safe, production-ready results with lifelike quality" and creative professionals have greater control over "structure, style, camera angles, and zooms while enabling rapid iteration in up to 2K resolution for precision cropping, recomposition, and large-format printing." Adobe says the new models mean it is easy to move from prompt-based AI image editing to Photoshop's traditional layers, masks, and selection tools. "Think of a portrait that needs a touch of magic," says Adobe. "Use Gemini 2.5 Flash Image Model (Nano Banana) to introduce floating floral elements while keeping the subject pristine. Refine with Photoshop's masks, curves, and color adjustments so those flourishes blend seamlessly into the shot." I ran a quick test of this feature, and I was impressed with the results. There are a few AI artifacts in there, but they could likely be fixed with some more editing. Gemini 2.5 Flash Image Model (Nano Banana) and FLUX.1 Kontext [pro] is available now in the Photoshop beta app. You may need to update to gain access to the new models.

Adobe expands Photoshop and Firefly Boards with new generative AI tools and third-party models - SiliconANGLE

Multinational creative software company Adobe Inc. is updating its suite of products with more artificial intelligence models to support professionals where they work. The company announced today that Firefly Boards, an AI-native software workspace for users to explore, iterate and collaborate in real time, will receive new AI generative video models including Runway AI Inc.'s Aleph and Moonvalley AI's Marey. Photoshop, Adobe's flagship image editor, is also gaining access to Google LLC's Gemini 2.5 Flash Image, nicknamed "Nano Banana," and Black Forest Labs Inc.'s FLUX.1 Kontext Pro. Adobe said it is rolling out Firefly Boards worldwide with new features such as presets, generative text edit, and describe image. Presets allow users to generate images in different styles. For example, "Product and Character" helps visualize products using image generation, while "Virtual Try On" lets users apply clothing to a model in a photo. Now in beta mode, generative text edit allows users to swap text directly in visuals, maintaining the style and context of the image. For example, a user might want to change the text of a neon sign or words on a billboard. Using text edit allows them to change the text, while maintaining the same "look" of the original neon letters or keep the font, structure, color and orientation on a sign. "Using Firefly Boards, I can quickly generate and iterate multiple visual concepts for a creative commercial proposal," said Malick Lombion, filmmaker and creative director. The new describe image functionality allows users to generate a ready-to-use natural language prompt that can be used as text for image generation. Users can edit the prompt afterwards to fit their needs. For example, a user could take an image of a brightly-colored bird and transform it into a text prompt, which they could then edit to change its setting, coloration, species or any number of other qualities and then have an Adobe image model render an image based on the new text. Photoshop already includes Adobe's Firefly AI models, which are commercially safe for use in creative workflows. The addition of Gemini 2.5 Flash Image and FLUX.1 Kontext broadens what creatives can do with generative fill, a feature that allows users to add, remove or transform parts of an image using natural-language prompts. "This represents a first in Creative Cloud applications to integrate partner models," Joel Baer, director of product management at Adobe told SiliconANGLE in an interview. "This is the beginning of our journey of integrating these models into Photoshop." Gemini 2.5 Flash Image excels at stylized elements, graphic details and scene additions while staying context-aware. It is also more consistent at making subtle edits without unintended changes. FLUX.1 Kontext is tuned for contextual accuracy, perspective, and environmental harmony. It performs well with complex tasks such as adjusting settings, working with human anatomy or transforming backgrounds while preserving core elements of a scene. When using these models, Photoshop automatically generates non-destructive layers that record both the prompt and the model used. This allows creatives to refine their edits, apply masks and separate AI-generated changes from the main image. For example, a designer could use Flash Image to add props or change a wine glass into a champagne flute, while FLUX.1 could transform an office desk from wood to glass without disturbing the items on top. "Many of our customers are using third party services in in combination with Photoshop as part of their creative process, whether it's for ideation or even editing images," said Baer said. He added that integrating these new models will open up more options for users in their creative flow. "Photoshop has always kind of been the place you go when you're not maybe totally sure what you need to do to your image, but you know Photoshop can do it." Adobe said for a limited time during the beta the new models, Image Flash and FLUX.1, will be available to everyone with a subscription and not deduct AI credits. Subscribers to standard plans will receive 100 lifetime generations per model and existing premium subscribers will get 500 generations per day per model.

Adobe Firefly App Now Lets You Generate and Edit Images Using Nano Banana

Adobe Premiere app for iOS was recently introduced by the company Adobe Firefly app is the newest home for Google's Gemini 2.5 Flash Image model. On Thursday, the software giant announced that the image generation and editing model, also known as Nano Banana, was being integrated into the Firefly text-to-image module within the app. The artificial intelligence (AI) model is a premium offering and will require a subscription to the platform. Apart from the Firefly app, Nano Banana has also been added to Firefly Boards, Photoshop's Generative Fill, and Adobe Express. Adobe Brings Gemini's Viral Nano Banana AI Model In a post, the software giant announced the integration of the Gemini 2.5 Flash Image model to its catalogue of third-party image generation models. Nano Banana now joins the likes of Google's Imagen 3 and 4, OpenAI's GPT Image, Runway Gen-4 Image, Flux 1.1 Ultra, and others. To use the Gemini model in the Firefly app, users will need to open the app, go to the Text-to-image option, and then select Nano Banana from the model picker. Once selected, they can simply add a prompt and generate their desired image. Additionally, users can also upload an image in the prompt bar and make edits to it. Since Flash Image is known for character consistency, users can edit different elements while keeping the rest intact. The image generation model also supports follow-up queries, which allows users to make multiple edits. Google's Nano Banana AI model recently went viral for its character consistency while editing images. Users have been reimagining their images as an action figure and in retro style over the last month. Now, those interested in using the model's capabilities in more creative and professional projects can directly jump to the Adobe Firefly app and begin creating and editing. Notably, the Firefly feature is a premium offering that requires credits. While free-tier users get limited monthly credits, the Adobe Firefly Standard subscription comes with 2,000 credits monthly. It is the cheapest subscription tier which costs Rs. 797.68 a month (inclusive of GST). It also allows subscribers to generate up to 20 five-second-long videos, translate up to six minutes of audio, and access Firefly Boards.