🚀 Midjourney Office Hours Highlights: November 27th

Midjourney is working hard to make their upcoming video and image model outstanding. The new video model, inspired by the V6 image model, and the enhanced V7 image model promise to elevate video and image generation capabilities. Alongside these advancements, Midjourney is developing world models, personalization features, and storytelling tools to enhance user experience.

For a quick overview, head straight to the TL;DR; section otherwise keep on reading.

Extended Recap

🎥 Video and Image Model Innovations

Midjourney is on the brink of unveiling groundbreaking advancements in both video and image generation. The team is diligently working on a new video model, which will echo the capabilities of the V6 image model but with a focus on video creation. Meanwhile, the V7 image model is poised to introduce a suite of new features and enhancements, promising a significant leap forward in image generation technology.

🌍 World Models and Future Releases

The development of world models is taking center stage, with data collection slated for December and training set to begin in January. This initiative is expected to yield exciting new features by February. The release of the video and V7 models is anticipated in January, with preliminary testing involving guides and moderators scheduled for December. The team is mindful of the holiday season and aims to maintain a balanced workload.

🖌️ Editor and Personalization Enhancements

Midjourney is enhancing its editor with a new version of the retexture mode, offering users more precise control. Quick improvements to moderation features are also under consideration. Additionally, the introduction of mood boards and personalization profiles is imminent, allowing users to curate multiple profiles and mood boards for a more tailored experience. This feature simplifies personalization by enabling users to incorporate multiple images seamlessly.

📖 Storytelling and Future Tools

A new storytelling tool, focused on world-building rather than traditional narratives, is set to launch in about two weeks. This tool will benefit significantly from the V7 model, offering higher quality images and the ability to depict multiple characters within a single scene. Looking ahead, Midjourney is also developing enhanced batch processing capabilities for the V6 model, potentially labeled V6.2, although this may be delayed to prioritize personalization features.

🔄 Creation Flow and Style Exploration

Efforts are underway to refine the creation process, making it more powerful and efficient. These enhancements will be integrated into the V7 model and future video functionalities. Additionally, plans are in place to develop an improved style explorer, aiming to make style exploration more accessible and free for users. This initiative will leverage spare computing resources to generate styles for community exploration.

🏢 Organizational Developments and Goals

Midjourney is restructuring its development process to increase the frequency of major model releases, aiming to boost the number from two per year to six or eight. This involves establishing a new data team focused on dataset iterations and an architecture team dedicated to model architecture improvements. The company is also exploring undisclosed hardware and software projects, with upcoming blog posts set to discuss psychometrics and demographics.

💻 Additional Projects and Tools

The development of a desktop application is under consideration, offering benefits such as standalone usage, job completion notifications, and file syncing. This approach is deemed more feasible than mobile apps due to app store restrictions.

TL;DR:

🎥 Video & Image Models: New video model akin to V6; V7 image model with major updates.

🌍 World Models: Data collection in December, training in January, features by February.

🖌️ Editor & Personalization: New retexture mode, mood boards, and profiles for personalization.

📖 Storytelling Tool: World-building focus, enhanced with V7, launching in two weeks.

🔄 Creation Flow: Improved batch processing and style exploration in development.

🏢 Organizational Goals: Increase model releases, new data and architecture teams.

💻 Additional Tools: Desktop app in consideration for standalone use and file syncing.

Stay tuned for more updates as Midjourney continues to innovate and enhance its platform!

If you want to support me, feel free to buy me a coffee ☕️ 

If you’re not subscribed yet to my newsletter Imagine Weekly, I’d be thrilled to welcome you on board!