A new system that produces images from text has arrived: “FLUX.1”

earthpressnews

A new system has arrived that generates images from posts. The name of this system is FLUX.1 and the artificial intelligence initiative behind it Black Forest is located.

Stable Diffusion A Germany-based company founded by researchers who developed the technology Black Forest‘s hand FLUX.1cloud based It can be tested on artificial intelligence systems or on computers with the developer version. 23 GB developer model Here The version included can produce very realistic results from what is written and It does not have problems with human hands and fingers like some of its counterparts. The system, which those who had the chance to test it found results at the Midjourney 6 level, already promises a great future. More sample images produced with the system Here located, company is also working on an OpenAI-backed Sora competitor that generates videos from text. Before this, the new version of Midjourney made a splash. Midjourney, a productive AI-based content creation system that visualizes what you write or in other words, the commands you give incredibly well, looks ambitious with its newly released V6.1 version.

YOU MAY BE INTERESTED IN

About this release, the company said, “V6.1 greatly improves image quality, consistency, and text reproduction, and comes with brand new upgrade and customization models. It’s smarter, faster, clearer, and more beautiful. We hope you enjoy our best model yet.” V6.1, which has better reproduction of hands, arms and legs than before, creates fine details better and visibly improves quality by reducing pixel artifacts.

Midjourneythe same character in different images before made a splash with its ability to use it. The service has been creating a character for a while, as you can see in the examples at the bottom of the page, can add different images without disturbing the overall structure. This was a feature that had been requested for a long time and truly started a new era in the service. The system, which can include the specified character’s facial features as well as hair, beard and clothing in different results, is currently in the first test phase, meaning it may reveal some minor errors or differences. However, the infrastructure, which is generally satisfactory to those who have tried it, opens very big doors and brings new content production opportunities.



lgct-tech-game