Imagen 4

4 · Family: Imagen

Google DeepMind's text-to-image model (2025). Three variants: Fast, Standard and Ultra; up to 2K resolution, improved in-image text rendering and typography, SynthID watermark.

✓ Active✓ Public accessImage generation📁 Imagen

Release date

20 May 2025

🏢Google DeepMindProducer 🏢GoogleProducer

Access:APIHostedDeployment:☁ Cloud

Overview

Imagen 4 is a text-to-image generation model developed by Google DeepMind, unveiled in 2025 (Google I/O) and subsequently made generally available in the Gemini API and Vertex AI. It belongs to the Imagen family and is its fourth generation.

The model comes in three variants: Imagen 4 Fast (fastest, cheapest), Imagen 4 (Standard, balanced) and Imagen 4 Ultra (highest quality and best prompt adherence). It generates images at up to 2K resolution, in 1:1, 3:4, 4:3, 9:16 and 16:9 aspect ratios. Compared with predecessors it improves detail, colors, styles and in-image text rendering and typography (e.g. captions, posters, packaging).

Imagen 4 is available through the Gemini API (model IDs imagen-4.0-*), Google AI Studio, Vertex AI and Google apps (Gemini, Whisk). All generated images include an invisible SynthID watermark. Gemini API pricing: $0.02 (Fast), $0.04 (Standard) and $0.06 (Ultra) per image.

Classification

Image generation

Family: Imagen

Applications

Image generation Creative advertising Social media content Personal visual expression Content generation

Access & deployment

APIHosted

Cloud

Weights: Closed

Key parameters

📥 Input: text

Technical specification

License

Wlasnosciowa (zamkniete wagi)

Hardware requirements

Cloud-only access (Gemini API / Vertex AI); no downloadable weights.

Modalities

⬇ Input

text

⬆ Output

image

Capabilities and applications

Native model capabilities

Text-to-image generation

A model's ability to create images from a text description (prompt), including control of style, composition, aspect ratio and rendering text within the image.

Category: vision

Application domains

Image generation Creative advertising Social media content Personal visual expression Content generation

Pricing

Technical architecture

Core Architecture

DMDiffusion Model

Deployment and security

🔒 Security / Enterprise

✓ Verified enterprise information

Updated: 30 Jul 2026↗ Security documentation

Sources and related pages

3 sources

WebGoogle DeepMind - Imagendeepmind.google DocsGemini API - Imagen (dokumentacja: warianty, 1K/2K, aspect ratio, SynthID)ai.google.dev DocsGemini API - cennik (Imagen 4 Fast/Standard/Ultra)ai.google.dev

Browse related topics

📁 Imagen 🌐 Image generation 🌐 Creative advertising 🌐 Social media content 🌐 Personal visual expression 🧠 Diffusion Model All image generation model models