Midjourney, accessible at https://www.midjourney.com/, has officially announced the alpha release of its latest image generation model, V7, for testing by the AI community. This new model boasts significant improvements in understanding text prompts, image quality, and feature coherence.

According to Midjourney, “V7 is an exceptional model, demonstrating enhanced intelligence with text prompts, producing fantastic image prompts, and noticeably higher image quality characterized by beautiful textures and improved coherence in details such as bodies, hands, and objects of all kinds.”

A key innovation introduced in V7 is the default activation of model personalization. Initially, users must take approximately five minutes to unlock this feature. Personalization can be toggled on or off at any time and is designed to substantially enhance the AI’s ability to interpret user preferences and aesthetic desires. Midjourney believes this feature sets a new standard for understanding user intent.

In conjunction with the V7 image generation model, Midjourney is introducing a feature called ‘Draft Mode.’ This mode promises to generate images at a rate ten times faster and at half the cost, significantly expediting the creative process.

The increased speed enabled by Draft Mode has allowed Midjourney to implement a unique “conversational mode” on its web interface. Users can instruct the system to make specific changes, such as replacing a cat with an owl or changing the time of day to nighttime, and the AI will automatically adjust the prompt and initiate a new image generation task.

Draft Mode also includes voice input functionality. By pressing the microphone button, users can verbally express their ideas and observe the generation of images in near real-time:

Midjourney views Draft Mode as an unprecedented method for refining creative concepts. If a generated image is appealing, users have the option to select ‘enhance’ or ‘vary’ to re-render it at full quality. Although draft images are of lower quality compared to standard mode, their behavior and aesthetic characteristics remain consistent.

The V7 image generation model will initially be available in two speed modes: Turbo and Relax. The standard speed mode is undergoing further optimization and is expected to be released soon. Midjourney has noted that Turbo jobs will cost twice as much as standard jobs, while draft jobs will cost half the amount.

Regarding other functionalities, features like upscaling, editing, and retexturing will initially utilize the V6 model, with updates planned for future releases. The functionality for mood boards and SREF is currently operational, with performance expected to improve in subsequent updates.

Looking ahead, Midjourney has outlined an active development schedule, with new features anticipated every one to two weeks for the next 60 days. A significant upcoming feature will be a new V7 character and object reference capability.

Finally, Midjourney advises users that V7 is a completely new model with its unique strengths and potential weaknesses. They encourage experimentation and feedback on its capabilities, noting that it may require different prompting techniques compared to previous versions.

(Image credit: Midjourney)

See also: Tony Blair Institute AI copyright report sparks backlash

Want to learn more about AI and big data from industry leaders? Consider attending the AI & Big Data Expo, which will take place in Amsterdam, California, and London. This comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.