The emerging artificial intelligence company Midjourney has announced its new model for generating images using smart technology, enabling the creation of images with long claims and advanced levels of detail, as well as the ability to add text to the image.
This release comes at a time when the company is facing increasing pressure from competitors, such as DALL-E 3, in addition to free tools like Meta and services like Leonardo also providing significant control over image generation.
The company has emphasized that version 6 of its core model features improved precise claims, enhanced logical coherence, precise control over colors and shadows, in addition to providing text generation capabilities for the first time.
The new version can now recognize subtle differences in punctuation marks and language rules.
Midjourney stated that version 6 provides precise intertwined requirements, sequential logic link, long requirements, and enhanced model understanding.
The company highlighted the progress made in version 6 compared to the previous version 5.1 released in May 2023.
Version 5 paved the way for the more advanced version 6, known for its short claims, user-friendliness, and aesthetic improvements.
The ability to create text is one of the prominent features in version 6, although it is not the primary feature of the model. Midjourney indicates that it is still considered a secondary feature.
This feature sets Midjourney in direct competition with other reputable models such as Dall-E 3 and Ideogram, but Midjourney’s style in text production is unique and distinguished.
In its modest description, the company described this feature as manifesting in the simple ability to generate texts, stating: “You need to write your text contained in quotation marks.” The company prioritizes style and aesthetics over textual accuracy.
The sixth version is slightly slower and more expensive compared to the fifth version, however, the team confirms that they are working on speeding up the model over time.
Version 6 features enhanced improvements in “precise” and “creative” modes, leading to double the image accuracy.
These innovations offer users a variety of creative options, in addition to a wide range of editing media to change resolution, differences between generations, and the creativity extent in the model.
The updates demonstrate that Midjourney continues to develop its favored model by many, maintaining its leadership even in the face of challenges from competitors relying on their internal models and the famous stable widespread open-source model.
