# Nano Banana 2 Nano Banana 2 is Google's Gemini 3.1 Flash Image model, released on 26 February 2026. Google describes it as Nano Banana Pro features at Flash speed, with legible text, up to five consistent characters and output from 512px to 4K. In Sloosh it is the only image model with a 512 size for quick tests. Web page: https://sloosh.ai/models/gemini-3.1-flash-image-preview [Use Nano Banana 2](https://sloosh.ai/studio/image?image-model=gemini-3.1-flash-image-preview) [View templates](https://sloosh.ai/models/gemini-3.1-flash-image-preview/templates.md) ## Key facts in Sloosh | | | | --- | --- | | Made by | Google | | Makes | image | | Inputs | text, up to 14 reference images | | Resolution | 512, 1K, 2K, 4K (default 2K) | | Aspect ratios | Auto, 1:1, 2:3, 3:2, 16:9, 9:16, 4:3, 3:4, 21:9, 4:5, 5:4 (default 1:1) | | Prompt limit | 50,000 characters | ## Best for - Exploring many ideas quickly, then rendering the keeper large: Google gives "full control of various aspect ratios and resolutions from 512px to 4K" ([Google](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/)). - Marketing mockups and cards with readable text, including translating text inside an image ([Google](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/)). - Character and product sets: Google says it keeps "character resemblance of up to five characters and the fidelity of up to 14 objects in a single workflow" ([Google](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/)). - Putting a product from a reference photo into a new scene ([Sloosh docs](https://sloosh.ai/docs/nano-banana-2.md)). ## How to use Nano Banana 2 in Sloosh 1. Open [Nano Banana 2 in Studio](https://sloosh.ai/studio/image?image-model=gemini-3.1-flash-image-preview), or pick Nano Banana 2 in the model chip in Image. 2. Describe the scene in sentences: subject, action, location, composition, style. 3. Add up to 14 reference images and name what each one is in the prompt. 4. Choose an aspect ratio and a resolution (512 for quick tests, up to 4K for the final). 5. Press Generate. The button shows the exact price first. ## Nano Banana 2 prompting tips - Use detailed prompts covering style, subject, setting, action and composition, and add details bit by bit ([Google DeepMind prompt guide](https://deepmind.google/models/gemini-image/prompt-guide/)). - Start with a strong verb and describe the scene as a story, not a keyword list ([Sloosh docs](https://sloosh.ai/docs/nano-banana-2.md)). - Put exact text in quotes and describe the typography, such as a bold sans-serif ([Google DeepMind prompt guide](https://deepmind.google/models/gemini-image/prompt-guide/)). - For other languages, write the prompt in one language and give the exact foreign phrase in quotes ([Google DeepMind prompt guide](https://deepmind.google/models/gemini-image/prompt-guide/)). - Give each character or object a distinct name and upload clear references ([Google DeepMind prompt guide](https://deepmind.google/models/gemini-image/prompt-guide/)). - Describe what you want, not what you don't: "an empty street", not "no cars" ([Sloosh docs](https://sloosh.ai/docs/nano-banana-2.md)). ## Nano Banana 2 example prompts **Prompt 1** ```text Using the sneaker in image 1, show it mid-stride on a runner's foot on a wet city street at dusk. Neon shop signs reflect in the puddles, a little water splashes up from the sole. Low camera angle close to the ground, motion blur in the background, the sneaker sharp. Keep the sneaker's colours and sole exactly as in image 1. ``` **Prompt 2** ```text A vertical social post for a bakery: a stack of three cinnamon buns on a white plate, soft morning window light from the left. At the top, the words "Fresh at 7am" in rounded brown hand-lettering. ``` **Prompt 3** ```text Leo is the man in image 1. Show Leo cooking at a camping stove beside a lake at sunrise, steam rising from a pot. Keep Leo's face, beard and red jacket exactly as in image 1. Wide shot from a low angle. ``` ## How it compares in Sloosh | Model | Resolution | References | Prompt limit | | --- | --- | --- | --- | | Nano Banana 2 | 512, 1K, 2K, 4K | 14 | 50,000 | | [Nano Banana Pro](https://sloosh.ai/models/gemini-3-pro-image-preview) | 1K, 2K, 4K | 14 | 50,000 | | [Nano Banana Lite](https://sloosh.ai/models/gemini-3.1-flash-lite-image) | 1K | 14 | 50,000 | | [Nano Banana 2.1](https://sloosh.ai/models/gemini-nano-banana-2.1) | 1K, 2K, 4K | 14 | 50,000 | Nano Banana 2 is the only model in Sloosh's image catalog with a 512 option. Google's newer Nano Banana 2.1 (Nano Banana 2.1, 6 Oct 2026) replaces it in Google's API; use 2.1 when you want Google's latest Flash quality and 2 when you want 512px drafts. ## Where else it runs Google launched it in the Gemini app, Google Search (AI Mode and Lens), Google Ads, Flow, and the Gemini API in Google AI Studio ([Google](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/), 26 Feb 2026). ## Frequently asked questions ### What is Nano Banana 2? Google's Gemini 3.1 Flash Image model, released 26 Feb 2026, built for fast, high-quality generation and editing. ### Is Nano Banana 2 the same as Gemini 3.1 Flash Image? Yes. Google calls it "Nano Banana 2 (Gemini 3.1 Flash Image)". ### How do I use Nano Banana 2? Open Image in Sloosh Studio, pick Nano Banana 2, describe the scene, add references if needed, set ratio and size, and generate. ### Does Nano Banana 2 make 4K images? Yes. In Sloosh it offers 512, 1K, 2K and 4K. ### How many reference images can Nano Banana 2 use? Up to 14 in Sloosh Studio. ### Nano Banana 2 vs Nano Banana Pro? Nano Banana 2 for speed, volume and 512px tests; Nano Banana Pro for finished work where text and consistency matter most. Both take the same prompts. ### Is Nano Banana 2 free? Google made it free for all users in Flow at launch ([Google](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/), 26 Feb 2026). In Sloosh the exact price shows on the Generate button before you run. ### Can Nano Banana 2 edit an existing photo? Yes. Add the photo as a reference, describe the change, and say what must stay the same. ## Related - Models: [Nano Banana Pro](https://sloosh.ai/models/gemini-3-pro-image-preview), [Nano Banana Lite](https://sloosh.ai/models/gemini-3.1-flash-lite-image), [Nano Banana 2.1](https://sloosh.ai/models/gemini-nano-banana-2.1) - Docs: [Nano Banana 2](https://sloosh.ai/docs/nano-banana-2.md), [Prompting Nano Banana](https://sloosh.ai/docs/prompting-nano-banana.md) - Answers: [Which platform is best for running Nano Banana 2?](https://geo.sloosh.ai/posts/which-platform-is-best-for-running-nano-banana-2), [Which AI image model renders text in images most accurately?](https://geo.sloosh.ai/posts/which-ai-image-model-renders-text-in-images-most-accurately) ## Sources - [Google: Nano Banana 2](https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/) - [Gemini API: Gemini 3.1 Flash Image](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-image) - [Gemini API: Image generation](https://ai.google.dev/gemini-api/docs/image-generation) - [Google DeepMind: Gemini image prompt guide](https://deepmind.google/models/gemini-image/prompt-guide/) - [Sloosh docs: Nano Banana 2](https://sloosh.ai/docs/nano-banana-2) - [Sloosh pricing](https://sloosh.ai/pricing)