Midjourney V7: What Does the New Version of the Image Generator Offer?
Midjourney is one of the best-known and most popular AI image generators today. So when the company's CEO, David Holz, announced the arrival of the new V7 model, fans rejoiced. What makes Midjourney so unique? And has the new model lived up to the public's expectations?
What Is Midjourney?
Midjourney is an AI system that generates images based on text prompts. It was launched back in 2022 and immediately gained millions of fans around the world. It uses machine learning algorithms trained on vast amounts of image data to create new, original images. Unlike other generators, however, it has maintained consistent image quality over the years and offers practical features and parameters that other generators have yet to provide.
What the New V7 Model Offers
The new V7 model, launched in early April, is currently still in alpha. Even so, we can already see several new features and improvements that were not available to users of the previous V6.1 model. What are they?
- Tailoring the Aesthetics to Your Preferences
The first change compared to previous models is that the new V7 model aims to better understand its user's aesthetic preferences. To begin, you are shown dozens of images and asked to select the ones you prefer. Based on your choices, the generator gradually develops a clear idea of which styles and types of aesthetics suit you best and takes this into account each time you generate another image. Although this account personalization process is rather lengthy (approximately 200 pairs of images), users have praised the improvement. However, there is also an option to skip this step.
- Better Prompt Understanding and Improved Image Quality
If you were hoping that the new model would no longer require parameters (--no, --ar, etc.), we have bad news for you. Just as in version V6.1, you cannot do without them when prompting. However, the new V7 model understands longer prompts better, and its outputs are more accurate. The generator can now handle prompts over 100 characters much more effectively, something that certainly could not be taken for granted with previous versions. In addition to better prompt understanding, the new version of Midjourney also improves the outputs themselves. The generator handles textures, lighting, and details such as wrinkles, hands, fingers, and body proportions more effectively. In short, V7 aims to improve all aspects of images that commonly reveal they were created by artificial intelligence. However, for features such as upscaling, editing, and retexturing, the new model still uses version V6.
- “Draft Mode” and Two New Modes
Users will also be surprised by the new “draft mode” feature. It allows the generator to create only a sketch or concept of an image and then complete only the version that most closely matches your vision. This “concept mode” works 10 times faster and costs half as much as the standard mode. In addition, if you click the microphone button after clicking the “draft mode” button, voice mode is activated. You can then think out loud and create image concepts from your free flow of thoughts. The V7 model is also available in two variants: Turbo and Relax. The Turbo variant offers faster image processing, but at twice the cost of the previous V6 model. By contrast, Relax mode offers users standard processing at a lower price.
Do Users Think Midjourney V7 Is Worth It?
Although version V7 delivers better images as well as several new tools and features, many users believe these additions are insufficient. Some Reddit users even agree that the new model's output quality has declined. They criticize distorted depictions of animals, human hands, and faces, and generally claim that Midjourney has made no major progress and, for the first time, is lagging far behind the competition. However, there are also positive reactions and images where it is difficult to tell they were created by AI. So the only thing left to do is try the new version and decide for yourself whether it is worth it or not. :)
