Google Launches Cheaper, Top-Ranked Image AI
Google's new image model, Nano Banana 2, is now the top-ranked model on Artificial Analysis benchmarks, according to industry podcasts. The model reportedly excels at complex edits like logo removal and generates accurate infographics, all while being "significantly cheaper" than previous pro versions. It's now live in Gemini and AI Studio, positioning it as a direct competitor for high-end creative work.
Nano Banana 2 combines the high-fidelity capabilities of Google's Pro models with the speed of its Flash models, making advanced image generation and editing faster and more accessible. This new version is rolling out across Google products, including Gemini and Search, and is available for developers through the Gemini API in Google AI Studio. All images generated by the model will include an invisible SynthID watermark to identify them as AI-created. The model introduces significant upgrades in text rendering, a historical challenge for AI image generators, now supporting accurate text in multiple languages. It can also generate a variety of logos and overlay them onto products or other surfaces. For developers, Nano Banana 2 offers new efficiency with a 512px resolution option to minimize latency and native support for additional aspect ratios like 4:1 and 8:1. On a cost-per-image basis, Google's models are positioned competitively against rivals like OpenAI. High-quality images from Google's Imagen 4 can cost as little as $0.06, compared to $0.167 for OpenAI's highest-quality equivalent. For subscription users, Google's AI Pro plan at $19.99 per month bundles image generation with other services like 2TB of storage, while OpenAI's $20/month ChatGPT Plus is more singularly focused on model access. Nano Banana 2 is part of the broader Gemini 3 model family, which emphasizes advanced reasoning and multimodal capabilities. This allows the image model to have a deeper understanding of prompts, leading to more logically consistent and realistic outputs. The underlying Gemini 3 Pro has shown more than a 50% improvement over its predecessor in solving benchmark tasks. For designers and agencies, these AI advancements are streamlining workflows by automating repetitive tasks like background removal and image resizing. This allows a greater focus on creative strategy and concept development. The ability to quickly generate multiple design variations and mockups accelerates client feedback loops and the overall design process. The integration of these powerful image models into platforms like Google AI Studio is changing how applications are built, enabling a "prototype-first" approach. Developers can use natural language to generate full app layouts and logic, with AI models for images, text, and search already connected, drastically reducing the time from idea to a shareable prototype.
