Midjourney is in the final testing stage of V7, a major update that overhauls almost every part of the AI image generator system.
In this week’s business hours session, developers explained that V7 has a new architecture, different datasets, and revised language processing methods.
Internal testing can last anywhere from a week to a month, followed by a two-week tuning phase and a potential alpha period with limited capabilities.
“The V7 model is a complete overhaul of V6, including a new architecture, different datasets, and revised language processing methods” Midjourney announced on its official server. “The alpha period usually follows an initial setup with limited features enabled.”
As the extensive redevelopment of V7 continues, the Midjourney team is exploring video creation opportunities through several approaches, including developing their own video model or partnering with external providers.
The main challenge, according to the team, is making video creation user-friendly and cost-effective.
Credit: Midjourney’s Discord posting.
The team hasn’t revealed a specific timeline for this video generator, so it could be in early development. For now, developers are working on “balancing speed, cost and product quality,” the statement said.
About 85% of users prefer photos created using the platform’s personalization system, which now includes mood boards and multiple profiles, Midjourney said. The company plans to expand these features by combining mood boards with style reference capabilities.
The platform also plans to introduce two different generation modes: a “real-time” option for similarly fast results Meta the “imagination” feature, Krea AI’s Doodle-to-Idage feature, or Leonardo’s real-time canvas.
They are also working in high-capacity mode to create large batches of images, similar to Leonardo’s Flow state.
Released in Midjourney V6 December 2023 and produced better results than its predecessor Midjourney V5.1 May 2023.
The results were noticeable, but the step from V5 to V6 wasn’t as big as the upgrade from V4 to V5. It was in the middle more cautious When talking about the V6, it’s reasonable to assume that the team was very pleased with the results before calling this new version a “complete overhaul.”
During this time, the team also announced its entry into the AI hardware market, but has yet to provide details.
Since the release of Midjourney V6, the team has expanded its capabilitiesadding features like image editor, custom interface, style customization tool, image scaling and resizing among others.
The team also began paying attention to user feedback, sharing various surveys to ask users what features they would like to see implemented.
Help us plan our roadmap for the next 12 months. Submit your idea here: https://t.co/TuxV2p6UyI and we combine your thoughts with ours and then create a voting board to understand what everyone thinks is most valuable. Thank you!
according to Encryption analysis, Midjourney currently has the edge in creative AI imaging, consistently delivering superior aesthetic results that stand out from its competitors.
Although it sometimes sacrifices tight speed control and text display accuracy, its versatility and overall capabilities make it a powerful tool.
The addition of customizable styles further enhanced the appeal of the platform by allowing users to create unique, personalized images that differed significantly from the platform’s signature image.
