Google has announced the general release of Imagen 4, the most advanced version of its AI model for converting text to images, making it available for free through AI Studio. However, it will only be free for a limited time.

In the official blog post, the company emphasized that Imagen 4 is a significant leap from previous versions, with improved accuracy, better and higher-quality content creation, and impressive results in displaying text within images – one of the most difficult technical problems for models of this type.
See also: Google adds AI capabilities to Chromebook Plus devices
Google introduced two variants of its new model:
- Imagen 4, designed for most everyday image creation, costs $0.04 per output image
- And Imagen 4 Ultra , the premium version of the model, which follows text descriptions ( prompts ) more accurately and offers even higher quality , priced at $0.06 per output image.
See also: The Browser Company: AI browser Dia in beta version
The company notes that the Imagen 4 Ultra performs better than competing models, especially in terms of compliance with complex text descriptions.
Imagen 4 is not currently available through Gemini, but the company has confirmed that it will be added soon for users with paid access.

Google's move to launch Imagen 4 through AI Studio with free access for a limited time is smart and strategically placed at a time when competition in the creative AI space has intensified significantly.
By providing free access to Imagen 4, Google creates a low "barrier to entry" for content creators, artists, developers, and marketers, attracting an audience that might not have tried it immediately if it were paid.
See also: OpenAI: ChatGPT Memory upgrade for “free users”
Google seems determined to seriously reposition itself in the generative visual AI. If Imagen 4 proves technically superior, and if the transition to commercial use is made with flexible pricing and proper integration into Gemini, then it could become a serious competitor to Midjourney, DALL·E, and even Stability AI.
Source: www.bleepingcomputer.com
