🚀 Midjourney Office Hours Highlights: January 15th

The Midjourney team is hard at work refining the much-anticipated V7 model, which promises to bring enhanced multilingual support, improved coherence, and better character references. However, the journey hasn't been without its hurdles. Data quality issues in the new dataset have caused delays, pushing the launch from January to February. The team is committed to ensuring that the V7 model meets high standards, even if some aspects of image quality and aesthetics may need further refinement beyond this release.

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

Extended Recap

🎥 Video Functionality: Quality vs. Speed

The team is exploring two main directions for video functionality: maximum quality and maximum speed. The former offers higher-quality output but is slow and expensive, while the latter is faster and cheaper but compromises on quality. This presents a challenge, especially for users on the $10/month tier. Midjourney is considering whether to develop an in-house video model or use external APIs, with plans to offer both fast and slow models to gauge user interest and usage.

🖼️ 3D Exploration: A New Dimension

There's renewed interest in 3D capabilities following promising internal tests with updated datasets. The team is assessing the feasibility of launching an initial 3D feature soon, though the level of polish and effort required remains a consideration. Long-term, there's potential for integrating 3D and open-world simulation experiences, which could revolutionize user interaction with the platform.

🔄 Big Batch V6 (V6.2): Enhanced Personalization

Before the V7 release, Midjourney might roll out an interim update, V6.2, allowing for larger batch sizes and improved personalization. Testing has shown significantly better user control within each generation batch, though final interface and UX details are still being refined.

👥 Character References and Omni-Reference

The V7 model will address previous inconsistencies in character references, such as hands and weapons. There's also potential for an "omni-reference" capability, allowing users to incorporate specific items like logos or coffee mugs. This reflects strong community feedback emphasizing the need for character consistency and style control.

💰 Cost, Server Usage, and Community Impact

Budget constraints limit the feasibility of extremely expensive model training, such as $100M+ for certain video models. However, large servers have been acquired in anticipation of new releases, allowing relaxed mode access for the $10 tier. The goal remains to keep releases financially sustainable, ensuring Midjourney can continue to be community-supported.

⚖️ Multiple Model Tiers: Fast vs. Slow

Observations show that most users prefer default or higher-quality modes, rarely opting for cheaper, lesser-quality options. Midjourney plans to explore whether a "fast" model for quick outputs or cheaper usage would gain traction. While professionals might pay more for higher quality, the broader community still desires affordability. Post-launch data will inform future development priorities.

🔮 Future Focus and Platform Vision

Midjourney anticipates rapid improvements in image quality, speed, and workflow over the next two years. They are considering adding advanced features like 3D-based exploration, more flexible style referencing, and community-driven style libraries. Balancing multiple projects, including V7, 3D, video, and style tools, is a challenge given the current team size.

🗣️ Community Feedback Priorities

Character creation, consistency, and advanced style controls are top community requests. There's also a desire for additional collaboration tools, such as user-generated styles. Updates to the Explore page and potential new style discovery features are in progress. The team plans to focus on V7 first, then finalize these additional initiatives as resources become available.

TL;DR

🚀 V7 Model: Launch delayed to February for data quality improvements; aims for better multilingual support and character references.

🎥 Video: Exploring high-quality vs. fast, cheaper options; considering in-house vs. external APIs.

🖼️ 3D: Promising tests; potential feature launch soon; long-term integration with open-world simulations.

🔄 V6.2: Interim release for larger batch sizes and improved personalization.

👥 Character References: V7 to improve consistency; potential for "omni-reference" capability.

💰 Cost & Community: Budget limits on expensive models; goal to remain community-supported.

⚖️ Model Tiers: Exploring fast vs. slow models; community desires affordability.

🔮 Future Vision: Rapid improvements expected; considering advanced features and balancing projects.

🗣️ Feedback: Focus on character creation, consistency, and collaboration tools; updates to Explore page underway.

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!