Midjourney, a generative image creation tool best known for running inside a Discord server, is spreading its AI wings. Midjourney’s creators announced Tuesday that they plan to introduce a “text-to-video” model in the coming months.
CEO David Holz said in an “Office Hour” Discord session that the company will begin training video models in January. This move represents a natural evolution for a platform based on a mature imaging model and stimulating competitive dynamics in the generative video industry.
Discord session notes include planned adjustments to V6 Niji, Midjourney’s comic/anime generator model, and consistency fixes for the upcoming official release of Midjourney V6. The company also wrote that its to-do list requires “starting training for new video models” and could be ready “within a few months.”
Neither Holz nor the Midjourney team shared any additional information about the model.
Midjourney is known for emphasizing quality and user experience over raw speed, even if it lags behind its competitors. The company has been rolling out enhancements like inpainting and outpainting, months after the feature went live on other platforms like Stable Diffusion, and its latest foray into rudimentary text generation is available on other models like the Dall-E 3, SDXL, or This was done after it became a common feature in . There are even less popular generators like Ideogram or IF.
Entering a crowded field
This venture into video came even after competing products were released. Stability AI recently announced Stable Video Diffusion. Meta just introduced its EMU video generator, and with established models like Pika and Runway ML marking the territory, Midjourney’s entry is shaping up to be a strong competitive landscape. Additionally, other image generators such as Leonardo AI have already implemented video generation capabilities, further intensifying the competition.
Midjourney’s recent v6 update boasts improved prompted follow functionality and more realistic images, and is the company’s latest effort to remain relevant and competitive. If your model shows some cohesion, even if it’s not yet perfect, you’ll have a solid foundation in the early stages.
The implications of these developments go far beyond corporate competition for hegemony. The creative and media industries are on the verge of transformation as Midjourney and others innovate and improve their products. The ability to create, manipulate, and interact with video content through AI opens up many possibilities, from making the jobs of celebrities and advertisers easier to potentially reshaping how we perceive reality.
Edited by Ryan Ozawa.
