Adobe Brings Google’s Gemini 2.5 Flash Image Into Firefly
The Photoshop maker adds the newest Gemini model to Firefly and Express, giving creators faster, sharper AI visuals with built-in safeguards
Adobe is expanding its Creative Suite with Google’s latest image-editing model. The company announced that Gemini 2.5 Flash Image is now integrated into Adobe Firefly and Adobe Express, giving creators a new way to generate visuals and bring their ideas to life.
To mark the occasion, Adobe is giving Firefly and Creative Cloud subscribers unlimited generations with Gemini 2.5 Flash Image through September 1. For new Firefly free plan customers, Adobe is providing up to 20 generations.
“With today’s addition of Google Gemini 2.5 Flash Image in Adobe Firefly and Adobe Express, people have even greater flexibility to explore their ideas with industry-leading generative AI models and create stunning content with ease,” Hannah Elsakr, Adobe’s vice president for new generative AI business ventures, remarks in a statement.
Google’s new model can be used in Firefly’s Text to Image module, Firefly Boards, and, as mentioned earlier, Adobe Express. It promises to provide finer control over editing photos, making higher-quality edits based on natural language requests. However, while we may celebrate the fact that we can now generate images that don’t distort the face, provide altered backgrounds, or disrupt the scene in unintended ways, this advancement doesn’t prevent others from using it for malicious purposes.
However, TechCrunch reports that Google feels Gemini 2.5 Flash Image provides the balance between security and creativity that people have been looking for. Images generated include Google’s SynthID watermarking system. Moreover, the company discloses that filtering and data labeling are used to “minimize harmful content in datasets and reduce the likelihood of harmful outputs.”
Be warned: While Gemini 2.5 Flash Image can create a wide range of images, it’s not without limitations. Specifically, Google stresses that creations may not be factually accurate (e.g., imperfectly rendered small faces, inaccurate spelling, and poor detailing), and character features may be incorrect.
Adobe has steadily opened Firefly to outside AI models, already supporting Google’s Imagen 4, Veo 2, and Veo 3, alongside options from OpenAI, Black Forest Labs, Runway, Pika, Ideogram, Luma AI, and others. But the addition of Gemini 2.5 Flash Image marks one of the fastest turnarounds yet for integrating a third-party model. The rapid adoption likely stems from two key factors: Adobe’s confidence in the mode’s compatibility with its Content Credentials watermarking system, and the belief that Gemini’s output meets the level of polish and professionalism its customers expect.