Another challenger to Midjourney

In addition to Flux, there is another tool for amazing AI Art.

Author

Erik Knobl
August 11, 2024

Good day, humans.

In today’s Newsletter:

Google’s Imagen 3 enters the competition with Midjourney and Flux

Google has opened Imagen 3, a Text-to-Image model developed by its experimental lab DeepMind, for public use through its ImageFX platform. The new AI Art tool is now available via ImageFX.

Imagen 3 launches with the ability to generate high-quality images with enhanced details, lighting, and prompt comprehension. It’s worth seeing how it compares to the current leaders, Midjourney and Flux:

The word "Light" made of various colorful feathers on a black background.

Cinematic photo of a beautiful young woman in sport shorts and shirt sitting on the floor of a big modern kitchen, with morning light.

Amigurumi elephant walking in the savannah, blurry background.

Why it matters

Imagen 3 has just been launched, but it already challenges Midjourney in terms of quality, and over time, it may match the versatility of the current leader in generative art. More competition and more options are always good for users.

Flux is now on Freepik

Just last week, we discussed the launch of Flux and its impact as a competitor to Midjourney. Almost immediately, Freepik announced its integration into their platform for premium users, and as of today, it’s already available.

Another platform where Flux is available is Flux AI Image Generator.

OpenAI continues to lose talent

Three OpenAI leaders are stepping down from their roles. OpenAI’s president, Greg Brockman, will also take a leave of absence for the rest of the year. Another co-founder of OpenAI, John Schulman, will leave the company to do more “hands-on work” on alignment at Anthropic, a direct competitor to OpenAI. Finally, ChatGPT’s product lead, Peter Deng, is also on his way out.

Google acquires Character AI’s technology and talent

Character AI is one of the most popular consumer AI products, with over 100 million monthly visits. And now Google has hired its founder and announced a deal to invest in the company, gaining access to its technology. It’s not an acquisition, but almost.

Runway: Turn an image into a video

Using Runway’s tool to generate a sequence from an image:

Original image

Generated video.

Midjourney: use this style

To use this style, add the parameter --sref 1806744320 at the end of your prompt. Example:

Photographic portrait --sref 1806744320

Thanks for reading.

See you next week! Hi 👋 I’m Erik Knobl, a Product Designer by day and a Generative AI explorer on weekends. I share my learnings in this newsletter. Consider subscribing to stay in touch.