--- title: "Nano Banana 2: Specs, Strengths & Prompt Tips | imaginary.app" description: "Nano Banana 2 (Gemini 3.1 Flash Image) is Google’s fast image model with legible text and output up to 4K. How it differs from Pro, prompt tips and limits." url: https://www.imaginary.app/models/nano-banana-2 --- Updated November 2026 # Nano Banana 2 (Gemini 3.1 Flash Image) Nano Banana 2 is Google’s Gemini 3.1 Flash Image model, launched February 26, 2026 to bring much of Nano Banana Pro’s quality and world knowledge to Flash-tier speed. Google recommends it for rapid generation and precise instruction following, and Nano Banana Pro for tasks that need maximum factual accuracy. Key takeaways - Official name Gemini 3.1 Flash Image; Google describes it as bringing Pro-level knowledge, quality and reasoning to “lightning-fast” Flash speed. - Generates legible text for mockups and cards, and can translate or localise text inside an image, per Google. - It is a thinking model: it plans the image before rendering, and Google’s docs say thinking cannot be switched off. - Google’s API docs now call it the previous-generation Flash image model and point new projects to Nano Banana 2.1. Google Generate with Nano Banana 2 Write a prompt, pick Nano Banana 2, choose your settings and see the credit cost before you start. Failed generations return their credits. [Open the generator](https://www.imaginary.app/generate)[View Pricing](https://www.imaginary.app/pricing) Sign in with Google to create an account; buy credits only when you need them. ## On imaginary.app These are the settings our generator offers for Nano Banana 2 right now, read from our live model catalog. - Type Image, from a text prompt - Resolutions 1K, 2K, 4K - Aspect ratios 16:9, 9:16, 1:1 - Starting price 0.64 credits (about $0.14–$0.16) at the lowest settings (1K, 16:9) The generator shows the exact credits for your settings before you start. USD ranges depend on which [pack or plan](https://www.imaginary.app/pricing) you bought credits with. If a generation fails, or you cancel it before it starts, its credits go back to your balance. Very rarely, when we can't confirm with the provider whether a generation finished, the credits stay reserved while we check, rather than being charged. Models other than Veo 3.1 need an account that has bought a credit pack or plan, and they have daily usage limits, per account and across the site: if one is reached, the generator tells you and you can try again the next day (limits reset at midnight UTC). ## The fast Gemini image model, not a cut-down one Nano Banana 2 sits in Google’s Flash line, which is tuned for speed and volume, but Google pitches it as carrying over the strengths of Nano Banana Pro: real-world knowledge, strict instruction following and clean text. When Google launched it, Google said Nano Banana 2 would replace Nano Banana Pro across the Fast, Thinking and Pro models in the Gemini app, and made it the default image model in Flow. That makes it the everyday Gemini image model, with Pro kept for harder jobs. ## It reasons before it draws Like other Gemini 3 image models, Nano Banana 2 is a thinking model: it plans the image before rendering it, and the docs state that thinking cannot be switched off. The practical effect is that it handles instructions with several conditions (“three jars, the middle one open, labels facing the camera”) more dependably when you spell them out, and it rewards prompts written as clear sentences rather than keyword lists. ## Text, translation and diagrams Google says Nano Banana 2 generates accurate, legible text for items such as marketing mockups and greeting cards, and uses Gemini’s knowledge base to render infographics, diagrams and data visualisations. Its docs list the languages in which image generation performs most reliably, including English, Spanish, French, German, Hindi, Japanese, Korean, Portuguese and Chinese, which makes it a sensible pick for non-English lettering. ## Nano Banana 2 specs from Google What Google and its API partners document for the model itself. A maker spec is not always a setting our generator offers: what you can choose on imaginary.app is listed in the block above. | Spec | As documented | Source | | ---------------------------- | ------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------- | | Release | Announced February 26, 2026 | [Google blog — Nano Banana 2 announcement, February 26, 2026](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/) | | Official model name | Gemini 3.1 Flash Image (\`gemini-3.1-flash-image\`), stable in the Gemini API | [Google AI Studio docs — Gemini 3.1 Flash Image model page](https://aistudio.google.com/docs/models/gemini-3.1-flash-image) | | Resolutions | 512px (0.5K), 1K, 2K and 4K | [Google AI Studio docs — Gemini API image generation](https://aistudio.google.com/docs/image-generation) | | Aspect ratios (fal endpoint) | Standard ratios plus extreme 4:1, 1:4, 8:1 and 1:8 | [fal — Nano Banana 2 API schema](https://fal.ai/models/fal-ai/nano-banana-2/api) | | Reasoning | Thinking model; thinking cannot be disabled (levels: minimal or high) | [fal — Nano Banana 2 API schema](https://fal.ai/models/fal-ai/nano-banana-2/api) | | Text | Legible in-image text; can translate and localise text within an image | [Google blog — Nano Banana 2 announcement, February 26, 2026](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/) | | Provenance | SynthID watermark on all generated images, paired with C2PA Content Credentials | [Google blog — Nano Banana 2 announcement, February 26, 2026](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/) | | Google’s positioning vs Pro | Nano Banana 2 for rapid generation and precise instruction following; Pro for maximum factual accuracy | [Google blog — Nano Banana 2 announcement, February 26, 2026](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/) | | Current status | Described in Google’s docs as the previous-generation Flash image model; Nano Banana 2.1 is recommended for new projects | [Google AI Studio docs — Gemini API image generation](https://aistudio.google.com/docs/image-generation) | ## Who Nano Banana 2 is for ### Good fit - Everyday images that need to follow a detailed instruction closely without waiting on a heavier model. - Marketing mockups, greeting cards and social graphics with short, legible text, including in non-English languages. - Simple infographics and labelled diagrams. - Infographics, diagrams and simple data visuals that draw on real-world knowledge, when speed matters more than Pro’s extra accuracy. ### Look elsewhere if - Text-dense infographics or long paragraphs where every fact and word must be right; Google steers those to Nano Banana Pro. - Photo editing, reference images or multi-turn refinement; on imaginary.app this model is text-to-image only. ## Prompt tips for Nano Banana 2 Be specific, and write full sentences as you would brief a designer. Google’s docs advise being “hyper-specific” (describe the armour, don’t just say “fantasy armor”), and the model’s reasoning step benefits from clear relationships between objects. Example prompt (written by us) ``` A cosy bakery window display at dawn: three tiers of croissants and cinnamon rolls on wooden stands, a chalkboard sign leaning against the glass, warm interior light spilling onto a wet cobblestone street, photographed from the pavement at eye level ``` Decide the exact copy first and put it in quotes. Google notes Gemini handles text more reliably when the wording is settled before the image is requested, so write the final words yourself rather than asking the model to invent them. Example prompt (written by us) ``` Birthday card design, watercolour wildflowers framing the edges, centred hand-lettered text "Happy 30th, Maya!" in deep plum, small line "with love from the studio" underneath, cream paper texture ``` Describe exclusions positively. Google recommends “an empty, deserted street with no signs of traffic” over “no cars”, because naming an object tends to invite it. Use camera language to control framing: wide-angle shot, macro shot, low-angle perspective. Google’s prompting advice lists these terms for controlling composition. Example prompt (written by us) ``` Macro shot of a single dew-covered spider web strung between two lavender stems, early morning backlight, soft lilac bokeh background ``` ## Limitations - Google’s search grounding, reference images and conversational editing are maker features that are not part of imaginary.app’s text-to-image integration; here the model works from your prompt alone. - Google’s docs note the model won’t always produce the exact number of images you ask for. - Every image carries Google’s invisible SynthID watermark. - Text outside the languages Google lists as performing most reliably may contain errors. - It is no longer Google’s newest Flash image model: the Gemini API docs recommend Nano Banana 2.1 for new projects. - Nano Banana 2 is made by Google; imaginary.app is not affiliated with Google. ## Related models - [Nano Banana Pro](https://www.imaginary.app/models/nano-banana-pro) - [Seedream 4.5](https://www.imaginary.app/models/seedream-4-5) - [FLUX Schnell (FLUX.1 \[schnell\])](https://www.imaginary.app/models/flux-schnell) - [Every model on imaginary.app](https://www.imaginary.app/models) - [AI image generator overview](https://www.imaginary.app/ai-image-generator) ## Frequently asked questions ### Nano Banana 2 vs Nano Banana Pro: what is the difference? Nano Banana 2 (Gemini 3.1 Flash Image) is the faster Flash-tier model. Nano Banana Pro (Gemini 3 Pro Image) is built on Gemini 3 Pro and uses heavier reasoning. Google recommends Nano Banana 2 for rapid generation and precise instruction following, and Pro for high-fidelity work that needs maximum factual accuracy, such as dense infographics. ### Is Nano Banana 2 free? In Google’s own tools, partly: Google made Nano Banana 2 the default image model in Flow, “available to all Flow users for zero credits”, and it powers image generation in the Gemini app under that app’s plans. On imaginary.app it uses credits. ### What resolution does Nano Banana 2 support? Google documents 512px, 1K, 2K and 4K outputs, and fal adds extreme aspect ratios such as 4:1 and 8:1. On imaginary.app the available resolutions and ratios are shown in the “On imaginary.app” panel. ### Is Nano Banana 2 the newest version? No. Google’s Gemini API documentation now describes Nano Banana 2 (Gemini 3.1 Flash Image) as the previous-generation Flash image model and recommends Nano Banana 2.1 for new projects. This page covers Nano Banana 2, the version offered here. ### Do Nano Banana 2 images have a watermark? Google states that all images generated by its image models include an invisible SynthID watermark, and pairs it with C2PA Content Credentials. ## Sources Prices, model specs and dates on this page were checked on November 10, 2026. Third-party prices change; check the source before you rely on a number. - [Google blog — Nano Banana 2 announcement, February 26, 2026](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/) - [Google AI Studio docs — Gemini API image generation](https://aistudio.google.com/docs/image-generation) - [Google AI Studio docs — Gemini 3.1 Flash Image model page](https://aistudio.google.com/docs/models/gemini-3.1-flash-image) - [fal — Nano Banana 2 model page](https://fal.ai/models/fal-ai/nano-banana-2) - [fal — Nano Banana 2 API schema](https://fal.ai/models/fal-ai/nano-banana-2/api) - [imaginary.app pricing](https://www.imaginary.app/pricing) How we checked: model specs come from Google's and its API partners' pages listed above; the imaginary.app settings and starting price come from our live model catalog. Example prompts are ours, not Google's. Commercial use: licence and usage terms for Nano Banana 2 are set by Google. Check them, and our [Terms of Service](https://www.imaginary.app/terms-of-service), before you publish what you make. Google ## Try Nano Banana 2 on your own prompt One account and one credit balance for Nano Banana 2, every other image model we offer, and Veo 3.1 video. The price shows before you generate. [Open the generator](https://www.imaginary.app/generate)[View Pricing](https://www.imaginary.app/pricing) Sign in with Google to create an account; buy credits only when you need them.