Image-focused AI models are now the primary driver of user growth for major AI applications, generating 6.5 times more downloads than traditional chatbot updates. According to a new report from app intelligence provider Appfigures, this marks a significant shift in how users engage with AI tools. Where conversational model upgrades and voice features once dominated user acquisition, visual capabilities have become the main reason people download and reinstall AI apps.

Why Are Image Models Outpacing Chatbot Updates?

The appeal of image generation is straightforward: it's tangible, shareable, and immediately impressive on a mobile device. Users can see results instantly and share them on social media, creating organic buzz that text-based improvements simply cannot match. When Google released its Gemini 2.5 Flash image model alongside the Nano Banana image capability in August 2025, the app saw more than 22 million additional downloads in the following 28 days, lifting overall installs by more than 4 times their baseline rate. Similarly, OpenAI's March 2025 launch of its GPT-4o image model generated more than 12 million incremental installs over 28 days, roughly 4.5 times the download activity associated with other model releases including GPT-4o, GPT-4.5, and GPT-5.

Even Meta's entry into visual AI demonstrated the trend. The company's September 2025 launch of its AI video feed Vibes added an estimated 2.6 million downloads over four weeks, showing that visual content, whether static images or video, consistently outperforms text-based features in driving user acquisition.

Does Download Growth Actually Translate to Revenue?

Here's where the story becomes more complex. While image models excel at bringing new users through the door, converting those users into paying customers is a different challenge. Appfigures found that download spikes do not automatically generate proportional revenue increases across all platforms.

Google's Nano Banana, despite driving the largest download spike at over 22 million incremental installs, generated only $181,000 in estimated gross consumer spending during its 28-day launch window. Meta's Vibes produced no meaningful revenue despite its user acquisition success. The exception was OpenAI: ChatGPT converted the attention from its GPT-4o image model into actual dollars, generating an estimated $70 million in gross consumer spending over the 28 days following launch compared with its prior baseline.

This disparity suggests that while image models are excellent acquisition tools, sustained monetization depends on factors beyond the feature itself, such as existing subscription infrastructure, user habits, and perceived value of premium tiers.

How Companies Are Leveraging Image Models for Growth

• Acquisition Strategy: Companies are prioritizing image and visual model releases as their primary lever for user growth, recognizing that these features generate significantly more downloads than conversational AI improvements or model parameter increases.
• Multimodal Focus: The shift reflects a broader industry movement toward multimodal AI experiences that combine text, images, and video, making products more engaging and easier to demonstrate on mobile devices.
• Monetization Packaging: Successful platforms like ChatGPT are bundling visual capabilities into subscription services and credit systems, converting initial user interest into recurring revenue streams rather than relying on one-time feature launches.

The data also reveals an interesting outlier: DeepSeek's R1 model drove 28 million downloads in January 2025, but this spike was driven by curiosity about the company's low-cost training methodology rather than any image capability. This highlights that while visual features are the most reliable acquisition lever, other factors like technological novelty and industry buzz can occasionally override the pattern.

Looking forward, the implications for product strategy are clear. As competition for user attention intensifies, AI companies are likely to tilt their roadmaps further toward generative visual features. However, the revenue lesson is equally important: downloads alone do not guarantee business success. Companies must pair visual innovation with clear monetization strategies, whether through subscriptions, credit systems, or cross-platform utility, to convert user interest into sustainable revenue.