# Google Imagen 4: specs, shutdown, and Nano Banana as the replacement

> Imagen 4 was shut down in August 2026; Google maps every Imagen model to Nano Banana 2.1. Imagen 4's resolutions and 5 aspect ratios, prices, and prompts that carry over.

URL: https://superduperai.co/blog/google-imagen-4/
Updated: 2026-11-02
Published: 2025-05-20

**Imagen 4 has been shut down.** Google turned off Imagen 4 Fast, Standard and Ultra in the Gemini API on 17 August 2026, and the docs now say "Imagen models are shut down. Use Nano Banana." Google's own deprecation table maps every Imagen 4 model to **Nano Banana 2.1**, which went GA on 6 October 2026. In the Gemini app, images come from Nano Banana 2.

So "Imagen 4 or Nano Banana?" has one answer now. Below is what Imagen 4 could do, since people still look it up, and how Nano Banana compares.

## Imagen 4 specs (as it was)

| | Imagen 4 Fast | Imagen 4 / Ultra |
| --- | --- | --- |
| Input | Text only | Text only |
| Max size | 1K (1024×1024, 1408×768…) | 2K (2048×2048, 2816×1536, 2560×1792…) |
| Aspect ratios | 1:1, 3:4, 4:3, 9:16, 16:9 | 1:1, 3:4, 4:3, 9:16, 16:9 |
| Images per prompt | up to 4 | up to 4 |
| Price (Vertex AI) | $0.02 | $0.04 / $0.06 |

There was **no 4:5**, no editing, no inpainting or outpainting, and no negative prompts. Prompts were English, with Chinese, Hindi, Japanese, Korean, Portuguese and Spanish in preview. Every image carried an invisible SynthID watermark.

**Timeline:** announced at Google I/O on 20 May 2025; preview in the Gemini API on 24 June; Fast, Standard and Ultra generally available on 14 August 2025; shutdown announced 15 June 2026; Vertex AI listed 30 June 2026 as its end date and the Gemini API turned it off on 17 August 2026.

## Imagen 4 vs Nano Banana

| | Imagen 4 | Nano Banana 2.1 |
| --- | --- | --- |
| Text to image | Yes | Yes |
| Edit your own image | No | Yes, also over several turns |
| Reference images | No | Up to 14 |
| Max resolution | 2K | 4K |
| Aspect ratios | 5 | More, including 1:4, 4:1, 1:8, 8:1 |
| Google Search grounding | No | Yes |
| API price per image | $0.04 (Standard) | $0.0336 (1K), $0.0504 (2K), $0.113 (4K) |

On the Artificial Analysis text-to-image arena (2 November 2026), Nano Banana 2.1 is #4, behind three OpenAI models; Imagen 4 is no longer listed.

## Using Nano Banana

- **Gemini app (free):** describe the image or attach a photo and say what to change. Free users download at 1K; Google AI plans get 2K and a "Redo with Pro" button for Nano Banana Pro. There is a daily limit that Google doesn't state as a number. Editing photos requires being 18+.
- **Google AI Studio:** free to try. The Gemini API itself has no free tier for images.
- **API prices per image:** Nano Banana 2.1 $0.0336 (1K), $0.0504 (2K), $0.113 (4K); Nano Banana 2 Lite $0.0336 (1K only); Nano Banana Pro $0.134 (1K–2K), $0.24 (4K). Batch is half price.

## Prompts that carry over

Google's Imagen guide is still a good base: **subject → context → style**, then refine one detail at a time ("a park in spring next to a lake" → "…the sun sets across the lake, golden hour" → "…red wildflowers"). Keep text short; Google suggested 25 characters or less for Imagen.

Nano Banana's guide adds: be specific ("ornate elven plate armor, etched with silver leaf patterns" rather than "fantasy armor"), say what the image is for, edit by conversation ("keep everything the same, but make the expression more serious"), build complex scenes step by step, and describe what you want instead of what to remove ("an empty, deserted street with no signs of traffic").

The first four examples are from Google's docs; the last one is ours.

Open: https://superduperai.co/library/nature-documentary-long-lens-prompt/

```text
A photorealistic wide-angle shot of a vibrant coral reef teeming with tropical fish. Crystal-clear turquoise water with sunbeams filtering down from the surface, illuminating a sea turtle gliding gracefully over the coral. Shot from a low perspective with a wide-angle lens. Aspect ratio 16:9.
```

```text
An icon representing a cute dog. The background is white. Make the icons in a colorful and tactile 3D style. No text.
```

```text
Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English.
```

Open: https://superduperai.co/library/style-lock-block/

```text
close-up photo of a woman in her 20s, street photography, movie still, muted orange warm tones
```

Open: https://superduperai.co/library/product-hero-shot-generation/

```text
Studio product photo of a matte white ceramic coffee cup on a light oak table, soft window light from the left, gentle shadow, a sprig of eucalyptus beside it. Minimal composition, neutral beige background, 4:3.
```

For OpenAI's image models see [GPT Image](/blog/gpt-image-1/); to animate a still, the [first-frame method](/library/image-to-video-first-frame-method/) and the [Veo 3 guide](/blog/veo3/).

## FAQ


### Is Imagen 4 still available?

No. Google shut down Imagen 4 Fast, Standard and Ultra in the Gemini API on 17 August 2026. Google's recommended replacement for every Imagen model is Nano Banana 2.1.

### Imagen 4 vs Nano Banana: which is better?

Nano Banana, and it is the only one still available. It adds editing of your own images, up to 14 reference images, 4K output and Google Search grounding, and at 1K it costs less per image than Imagen 4 Standard did.

### What resolution and aspect ratios did Imagen 4 support?

Imagen 4 and Imagen 4 Ultra went up to 2K (for example 2048×2048 or 2816×1536); Imagen 4 Fast stopped at 1K. Aspect ratios were 1:1, 3:4, 4:3, 9:16 and 16:9. There was no 4:5.

### Is Nano Banana free?

In the Gemini app and Google AI Studio, yes, with a daily limit. Free Gemini users download at 1K. The Gemini API has no free tier for images.


## Sources

Checked on 2 November 2026:

1. [Gemini API: Imagen](https://ai.google.dev/gemini-api/docs/imagen) (shutdown notice)
2. [Gemini API: deprecations](https://ai.google.dev/gemini-api/docs/deprecations)
3. [Gemini API: Nano Banana image generation](https://ai.google.dev/gemini-api/docs/image-generation)
4. [Gemini API: changelog](https://ai.google.dev/gemini-api/docs/changelog) and [pricing](https://ai.google.dev/gemini-api/docs/pricing)
5. [Vertex AI: Imagen 4 model page](https://cloud.google.com/vertex-ai/generative-ai/docs/models/imagen/4-0-generate) and [pricing](https://cloud.google.com/vertex-ai/generative-ai/pricing)
6. [Vertex AI: Imagen prompt guide](https://cloud.google.com/vertex-ai/generative-ai/docs/image/img-gen-prompt-guide)
7. [Gemini Apps Help: images](https://support.google.com/gemini/answer/14286560?hl=en)
8. [Artificial Analysis: text-to-image](https://artificialanalysis.ai/image/leaderboard/text-to-image)
