We’ve been coexisting with AI-generated images for quite some time now, but this week, several major players made significant leaps forward. Specifically, I’m referring to substantial updates in Midjourney, Google’s new model, and Grok.
Contents
- Midjourney hits the web
- Grok-2 unleashes the monster
- Google gets competitive with Imagen 3
Each company showcases the technology evolving at varying paces and in diverse directions. It’s still very much an open arena, and each enterprise demonstrates precisely how far the advancements have reached.
Recommended Videos
Midjourney hits the web
An AI image generated in Midjourney.
Channel/Midjourney
Let’s commence with Midjourney, which quietly introduced a new web editor late Thursday. This editor integrates a number of useful image manipulation tools into a solitary user interface.
Previously, functions such as reframing, repainting (adding AI-generated assets to or modifying an existing image), panning, canvas extension (expanding the boundaries of the image and generating content to fill), and zooming all demanded their individual specific tool for usage and were situated across multiple menus, compelling creators to constantly switch back and forth. This novel UI offers a more coherent and streamlined editing process, representing a marked deviation from the program’s initiation on Discord.
Midjourney just released their web editor!!
It’s actually one of the coolest features they’ve dropped in a while
essentially lets you do inpainting, panning, zooming and more all in a single step
really really powerful pic.twitter.com/Wgyi9ElE5N
— Nick St. Pierre (@nickfloats) August 16, 2024
The new web editor is devised to render the editing of AI-generated images more effortless and seamless, as per Midjourney CEO David Holz on Discord recently. “We think this makes editing your MJ images way more seamless than before and is a huge step forward,” he wrote.
Although Midjourney persists in migrating away from Discord towards becoming a web-based application, the company also declared that it will mirror messages from popular channels like “daily-theme,” “prompt-craft,” and “general-1” between its web rooms and Discord channels, enabling people to follow those threads from whichever platform they prefer. The company also introduced a new selection tool that operates like a digital brush and has supplanted both the square selection and lasso tools.
The new editor is accessible to all Midjourney users who have already generated more than 10 images on the platform. The initial responses from the creator community have been largely affirmative.
The editor arrives two weeks after the release of Midjourney 6.1, which enhanced image quality and coherence (such as the correct number of fingers), as well as significantly improved processing times and understanding text accuracy in its image prompts.
Grok-2 unleashes the monster
The Midjourney update emerges just two days after the release of Grok-2 by Elon Musk’s xAI startup, which constitutes the next major event this week.
Grok’s image generation capabilities are empowered by the Flux.1 model from Black Forrest Lab, which has been rapidly gaining popularity due to its remarkable image quality and free usage.
24 hours since the launch of Grok 2.0 and its image creation capabilities!
I’ve prepared these 9 examples for you to make the most of it + access to an enless prompt library!
(Bookmark this for later) pic.twitter.com/7EDYSogfV2
— TechHalla (@techhalla) August 15, 2024
The most significant controversy surrounding Grok-2 isn’t merely its quality, which is rather good, but its seemingly ill-defined guidelines. Unlike many of the other AI image generators, Grok-2 appears to have very few guidelines regarding intellectual property, violence, and other explicit content. This isn’t the first instance where an AI image generator has encountered this type of blunder, but with Grok, it feels intentional, with Musk referring to it as “the most fun AI in the world.”
People have already tested its limits and created all sorts of dreadful and peculiar imagery, evoking the early days of AI image generation. But if you adhere to Musk’s rhetoric, Grok-2’s lack of guidelines seems purposeful and could ultimately shape the evolution of this technology in the future.
Google gets competitive with Imagen 3
An AI image generated by Google’s Imagen-3 model.
Google
Finally, Google announced its new Imagen 3 AI model, which was released to all U.S. users on Thursday. Google labels it as its “highest quality text-to-image model,” now capable of generating “better detail, richer lighting and fewer distracting artifacts than our previous models.” Google also asserts that Imagen-3 is superior at rendering text and now comes in different versions, tailored for the task at hand, such as something light like a quick sketch or something far more detailed and high-resolution.
For the time being, Imagen 3 is only accessible through Google’s AI Test Kitchen, as a component of ImageFX. This is currently in closed beta, meaning you’ll have to join the waitlist if you aren’t already a participant.