Imagen by Google
Imagen is Google DeepMind's text-to-image model, known for photorealistic image quality and strong rendering of text and typography inside images. The latest version, Imagen 4, generates images up to 2K resolution and is available in the Gemini app, Whisk, Google AI Studio, the Gemini API and Vertex AI.
Free to try in the Gemini app / Whisk; API $0.02/$0.04/$0.06 per image (Fast/Standard/Ultra).
How to use Imagen by Google
- Open image generation in the Gemini app, Google AI Studio or the Gemini API, where Google's image models now live — the Imagen page itself now presents Gemini Image, also called Nano Banana.
- Write a detailed prompt describing what you want to see — the characters, the setting and the overall feel — since more detail brings the image closer to what you imagined.
- Upload a reference image and add text instructions to create complex, detailed images, using everyday language and keeping the conversation going to refine the result.
- Review the generated image and continue the conversation to adjust it, relying on the model's real-world knowledge so the image follows real-world logic.
Frequently asked questions
Is Imagen free?
Image generation powered by Google's image models is available to free Gemini app users at no cost, with a Google account, though free use carries daily limits that Google does not publish as one fixed number.
What is Imagen best for?
Generating detailed, playful, realistic or whimsical images from text, including images with legible text, and editing images conversationally with a reference image.
Where can I use Imagen?
Inside the Gemini app, Google Pics, Google AI Studio and the Gemini API, as well as in Workspace apps such as Slides, Docs and Vids.
Pros & cons
Pros
- It generates detailed, realistic or whimsical images from plain-language prompts, with strong text rendering inside images.
- You can upload images and refine results conversationally, editing with everyday language instead of starting over.
- It is built into Gemini, Google AI Studio, the Gemini API and Workspace apps, so images can be made where you already work.
- Free Gemini users can generate images at no cost, with paid plans raising the allowance for heavier use.
Cons
- There is no separate Imagen website to use on its own any more — the Imagen page now directs you to Gemini Image (Nano Banana) inside Gemini.
- Free use is capped by daily limits that shrink when demand is high, so heavy image work needs a paid plan.
Related Image Generation & Photo tools
- MidjourneyPaid · ★4.8 · Midjourney
- DALL·E 3Freemium · OpenAI
- Leonardo AIFreemium · Leonardo AI
- IdeogramFreemium · Ideogram
- FLUXPaid · Black Forest Labs
- Stable DiffusionFree · Stability AI
- Adobe FireflyFreemium · Adobe
- Microsoft DesignerFree · Microsoft