🚀 Midjourney Office Hours Highlights: December 18th

In the latest Midjourney office hours, many exciting updates and future plans were unveiled, promising to enhance the creative experience for users. From innovative tools to upcoming model enhancements, there's a lot to look forward to in the Midjourney universe.

The new mood boards are an amazing addition to the Midjourney ecosystem go chek it out at https://www.midjourney.com/personalize.

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

Extended Recap

🌟 Recent Feature Releases

Midjourney has rolled out mood boards and multiple personalization profiles, allowing users to tailor their creative experiences more precisely. The team encourages the use of custom models to achieve superior results compared to baseline models. Additionally, an intriguing experimental research feature named "network" has been introduced, aimed at exploring world-building and storytelling.

🔍 Future Sharing and Exploration Tools

Looking ahead, Midjourney plans to expand the exploration of S-REFs and mood boards, facilitating broader community sharing. The team harbors a long-term vision of enabling aesthetic exploration that transcends all previous human history. They aspire to coordinate community creativity to reach an "aesthetic singularity," a point where artistic expression reaches unprecedented heights.

📸 Upcoming Batch Image Features

The team is contemplating "batch 8" features to manage larger sets of images effectively. However, the priority remains on mood board sharing before these batch features. There is an ongoing discussion about optimizing image resolution versus batch size, ensuring users get the best of both worlds.

🛠️ Version 7 (V7) Model Development

The development of the V7 model is in full swing, with a potential release by the end of January. This version focuses on character consistency and enhanced character references, with possible improvements in style and object references. The team is exploring optimal resolution, batch sizes, and upscaling strategies. Post-release, they plan to iterate quickly with follow-up models like 7.1, aiming for a steady and frequent release cadence by decoupling architecture, data, and scaling.

🎥 Video Model Considerations

Video models are currently yielding mixed results, with challenges in balancing quality, speed, and cost. Some video functionality is expected to be released by January to assess community interest. The team is evaluating whether to develop in-house video models, partner with third parties, or offer multiple options, acknowledging that high-quality video models may not yet be cost-effective.

🌍 World-Building and Interactive Features

Midjourney is keen on long-term projects involving world-building, storytelling, and immersive experiences. Future possibilities include "walk-around" or real-time interactive features, as well as comic book-like storytelling and scenarios integrating multiple characters.

🖥️ Infrastructure and Server Capacity

Thanks to early hardware arrivals, Midjourney currently enjoys a surplus of server capacity. They have introduced a "holiday relaxathon" period, offering relaxed mode to all users, with plans to reduce or eliminate relax mode wait times. This initiative aims to gather data on server usage and community interest in high-volume generation.

🔧 Emphasis on Fundamentals and Ongoing Improvement

The team acknowledges that many users primarily engage with basic features rather than advanced tools. There is a commitment to enhancing core aspects such as speed, resolution, quality, and prompt comprehension. Midjourney strives to balance new feature innovation with refining the basics.

🎨 Aesthetics and Beauty in Model Output

Midjourney is dedicated to developing models that not only produce realistic results but also create more beautiful images. Users are encouraged to personalize models for improved aesthetic outcomes. The team plans to devise methods to enhance overall visual appeal and engage more deeply with design elements.

📈 Data, Scaling, and Release Cadence for 2025

A new approach to model versioning is in the works, ensuring more frequent and focused updates. Multiple models following V7 will incorporate various improvements in data, scaling, and architecture. The goal is to continuously learn from community feedback and usage patterns to guide future directions.

TL;DR:

• 🌟 Recent Features: Mood boards, personalization profiles, and the "network" feature for storytelling.

• 🔍 Future Tools: Enhanced sharing of SREFs and mood boards, aiming for an "aesthetic singularity."

• 📸 Batch Features: Considering "batch 8" for larger image sets, prioritizing mood board sharing.

• 🛠️ V7 Development: Focus on character consistency, with a release expected by January.

• 🎥 Video Models: Mixed results; some functionality to be released by January.

• 🌍 World-Building: Interest in immersive storytelling and interactive features.

• 🖥️ Infrastructure: Surplus server capacity, "holiday relaxathon" for relaxed mode access.

• 🔧 Fundamentals: Commitment to improving speed, resolution, and quality.

• 🎨 Aesthetics: Focus on creating beautiful images and personalized models.

• 📈 2025 Cadence: Frequent updates with improvements in data, scaling, and architecture.

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!