Google has recently added image generation to Bard through their Imagen 2 text-to-image model. The service is completely free and allows for the creation of two images simultaneously.
The ImageFX AI tool, which is based on Imagen 2, has been developed with additional guardrails for safety. However, at present, the feature is only available for prompts in English and is not accessible in the UK, Switzerland, and EEA regions.
OpenAI introduced Dall-E 3 image generation for paid ChatGPT Plus members in October 2023, and Google has now followed suit. Although Google is a bit late to the party, it is using its Imagen 2 AI model to generate images using text prompts.
With new guardrails, Google has built the ImageFX tool based on the Imagen 2 model, and it’s integrated into Bard. Here’s how you can use it to create images in Bard:
- Visit bard.google.com on your desktop or smartphone browser.
- In the text field, start your prompt with “create an image of…” or “generate an image of…” and describe what you want to create. Currently, only prompts in English are supported.
- After a few seconds, Bard will generate two images. If you wish to create more AI-generated images, click on “Generate more.”
- Please keep in mind that the image resolution of the photo is 512 x 512 pixels, and it’s available to download in JPG format.
- Currently, it’s not possible to increase the size of these images generated by AI.
- Additionally, if you’re located in the US, you can access Google’s ImageFX tool directly by visiting AI Test Kitchen.
- Users under the age of 18 are not allowed to create images in Bard, and the tool may be renamed to Gemini in the future.
- Although Bard’s image generation capabilities are not as advanced as Midjourney’s and OpenAI’s Dall-E 3 model, it’s still a useful and free tool for creating images.
- Microsoft also offers a similar service through their Bing AI image creator, which is powered by Dall-E.