🚀 Midjourney Office Hours Highlights: January 22nd

The Midjourney team is really good at keeping us on our toes with the release date of V7. At least February is still on 👌 As the team gears up for the release of the V7 model, several exciting developments are on the horizon, promising to enhance user experience and expand creative possibilities.

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

Extended Recap

🎥 Video Model Development

Midjourney is gearing up to release two distinct video models in collaboration with partners. The primary aim is to assess community interest in models that are either fast and cost-effective or slow and premium-priced. This strategic approach allows Midjourney to test the market response before investing significant resources. The initial release will be experimental, providing valuable insights into user preferences. Additionally, Midjourney plans to enhance these partner models with their unique "Midjourney magic," ensuring a distinctive edge.

🖼️ V7 Model Progress

The development of the V7 model is back on track after overcoming data contamination challenges. Currently in the final tuning phase, the model is expected to launch tentatively by mid-February. Users can look forward to substantial improvements, including superior image quality compared to V6, better prompt understanding, and enhanced multilingual capabilities. The V7 model will also feature a more accurate character reference system and a potential "omni reference" feature for logos and objects. The team is diligently addressing bugs and optimizing the model in parallel to ensure a smooth release.

🔧 New Features Pipeline (30-60 Day Outlook)

Midjourney is actively working on several new features set to roll out in the next 30 to 60 days. Among these is the introduction of 3D capabilities, allowing for camera movement and reframing in 3D space, along with a rerender capability for polished versions. The batch mode is also being expanded to support larger batch processing and incorporate a user preference learning system, with early tests showing promising results. Post-V7, the focus will shift to developing both slower and faster model variants, with real-time V7 targeting better performance than V5. Additionally, improvements in upscaling for higher resolution are being explored.

🔮 Future Developments

Looking further ahead, Midjourney has multiple secret projects in the pipeline, with most expected to be announced within the next six months. One project is pending until an order placement capability is established. The team is also exploring new moderation systems and evaluating larger datasets for potential implementation. A key area of investigation is the speed-quality trade-off for real-time processing, with options ranging from 5 seconds to 30 times per second. The overarching goal is to maintain quality while enhancing speed, ensuring users receive the best possible experience.

TL;DR

🎥 Video Models: Two models in development to test community interest in fast/cheap vs. slow/expensive options, with "Midjourney magic" enhancements.

🖼️ V7 Model: Set for mid-February release, offering better image quality, prompt understanding, multilingual capabilities, and more.

🔧 New Features: Upcoming 3D capabilities, expanded batch mode, and post-V7 model variants focusing on speed and resolution improvements.

🔮 Future Plans: Secret projects in the works, new moderation systems, and exploring speed-quality trade-offs for real-time processing.

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!