# Nano Banana 2.1 Text to Image AI > Create an image from text with Nano Banana 2.1. Describe a scene, a poster or an illustration, choose the canvas shape, and generate without uploading a source picture. Web page: https://banana2.pro/nano_banana_2_1_text_to_image_ai ## Current settings and credits - Model: Nano Banana 2.1 - Price: 1K: 3 credits per image; 2K: 4 credits per image; 4K: 6 credits per image - Aspect ratios: auto, 1:1, 9:16, 16:9, 3:4, 4:3, 2:3, 3:2, 4:5, 5:4, 21:9, 4:1, 8:1, 1:4, 1:8 - Images per request: 1 - Text-only input; this entry point does not accept reference images. - Default resolution: 1K Review the total shown in the panel before submitting. The values above come from the current website configuration; examples below explain the workflow. ## Make a visual brief out of a written idea Nano Banana 2.1 text to image is a starting point for original compositions: a story illustration, a fictional room, a product concept or a simple announcement. It is most useful when you can explain what the viewer should see but do not yet have a photo or drawing that captures it. For text to image, describe the subject and its relationship to the space around it. A small bird on a broad windowsill tells a different story from a close-up of the same bird. Add style, light and color after the composition is clear. This gives the AI art generator a defined scene instead of a list of disconnected qualities. Nano Banana 2.1 offers standard and extended aspect ratios here, so you can write for a wide banner or a tall illustration from the beginning. Treat the shape as part of the creative brief. Stretching the idea of a square picture across a panorama often leaves the extra space without a purpose. ## How to write a Nano Banana 2.1 text to image prompt Use ordinary sentences to explain the subject, composition and message. ### 1. Describe what appears and where Start the text to image prompt with the main subject, the setting and the viewpoint. For example: “A tiny red cabin at the edge of a snowy forest, wide view, with most of the sky left clear.” Placement gives the result a structure you can evaluate. ### 2. Add style, lighting and exact short text Tell Nano Banana 2.1 whether the scene should be a photograph, a print illustration or a paper collage. If words belong in the picture, quote them and specify their location. Keep a headline separate from visual instructions so it is clear which text should appear. ### 3. Choose the format and inspect the result Pick the aspect ratio and resolution, then check the price before generating images from text. Review the subject, the balance of the composition and any lettering. Keep the useful parts of the brief and change the most important mismatch for the next attempt. ## Text to image prompts for different formats Adapt the composition to the output instead of adding more adjectives to the same prompt. ### A panoramic illustrated scene Try this Nano Banana 2.1 prompt: “A long coastal village at sunrise, simple painted shapes, boats along the lower edge, a wide open sky, muted blue and coral.” Choose a wide canvas for the text to image result and check that the horizon stays coherent across the frame. ### A vertical community poster Use the AI art generator for a poster direction: “A vertical seed exchange poster, large title ‘SMALL SEEDS, BIG IDEAS’, one illustrated sunflower, cream background, green lettering.” Keep dates and detailed instructions out of the first composition, then add verified final copy in a design tool. ### A room concept with a clear focal point Generate an image from text such as: “A reading corner with one ochre chair beside a tall window, a low shelf behind, soft afternoon light, no other furniture.” A limited object list makes it easier to judge layout and scale than a room full of competing details. ## Nano Banana 2.1 text to image FAQ ### Can Nano Banana 2.1 generate images without a photo? Yes. This text to image page needs only a written prompt and output settings. It creates one image per request. Use it for a new design or imagined scene; switch to an image-based page when a particular source picture should guide the result. ### What is a good Nano Banana 2.1 prompt? A useful text to image prompt explains the subject, placement, setting, style and lighting. Include only constraints you can inspect in the finished picture. “A chair on the right with open wall space on the left” is more actionable than “make a perfect composition.” ### Can Nano Banana 2.1 put words in an image? You can request short text, such as a headline or sign, and quote the exact wording. After text to image generation, inspect every character and its placement. Long paragraphs, dense data and exact fonts are better finished in a dedicated layout application. ### How much do 1K, 2K and 4K images cost? Nano Banana 2.1 uses 3 credits at 1K, 4 at 2K and 6 at 4K. The panel displays the total before submission. Each attempt to create an image from text is a new generation, even if you keep the same prompt. ### Can I make a very wide or tall image? Yes. Alongside common ratios, this text to image tool offers 4:1, 8:1, 1:4 and 1:8. Describe how the subject should occupy that shape. A panorama needs intentional spacing, and a very tall composition needs a clear vertical reading order. ### Can text to image reproduce the same character every time? A repeated description can guide appearance, but separate images from text may differ. Save an approved character image and use it on the [reference-based page](https://banana2.pro/nano_banana_2_1_image_to_image_ai) for the next scene. Inspect the new result rather than assuming the reference will preserve every feature. ### Does this page search the web for accurate information? This Nano Banana 2.1 interface has no web-search control. Supply the information that should appear in the picture and verify it yourself. A text to image result should not be treated as a researched answer, even when a diagram or sign looks convincing. ## Write for the picture you want to see Use Nano Banana 2.1 with a clear arrangement and a format that suits the idea. Generate a first version and revise the brief based on what the image actually shows. [Open the tool](https://banana2.pro/nano_banana_2_1_text_to_image_ai)