
Powered by Gemini 2.5 Flash, Nano Banana delivers fast photorealistic image generation from text prompts. The model behind the viral 3D figurine trend across social media.



Nano Banana — internally codenamed by Naina Raisinghani at Google DeepMind — is built on Gemini 2.5 Flash and optimized for photorealistic text-to-image generation. It first appeared anonymously on the Chatbot Arena leaderboard on August 12, 2025, before its public launch on August 26, 2025. Within days, its hyper-realistic 3D figurine output went viral across social media. Independent reviews from outlets like TechRadar noted that Nano Banana produces more realistic and internally consistent results than competing image generators, with particular strength in architectural visualization and correct spatial scale even with complex geometry.
Convert detailed text prompts into photorealistic images. Excels at 3D figurines, architectural scenes, and product shots with accurate spatial relationships and correct scale.
Five aspect ratios available: 1:1 (square), 16:9 (widescreen), 9:16 (vertical/stories), 4:3 (standard), and 3:4 (portrait) — covering every major platform format.
Export in PNG for lossless quality or JPEG for smaller file sizes. All generated images carry an invisible SynthID watermark for AI-content provenance.
Photorealistic images created with Nano Banana, showcasing its strength in figurines, portraits, and architectural scenes.








Describe the image you want in natural language. Include details about subject, lighting, camera angle, and style for best results.
Pick an aspect ratio (1:1, 16:9, 9:16, 4:3, or 3:4) and output format (PNG or JPEG) to match your use case.
Nano Banana processes your prompt and returns a photorealistic image. Download it directly — every output includes an invisible SynthID watermark.
Generate photorealistic AI images from text prompts. The same model that launched the viral 3D figurine trend — 5 credits per image.
2K cinematic video with native audio
1080p video generation with audio
4K video generation model
Video generation with audio support
Turbo Pro video generation
Next-gen AI image generation
4K AI image generation
AI image editing model
Ultra-fast AI image generation