Midjourney brings its signature look to video with V1

Midjourney just launched its first-ever video generation model, V1, bringing motion to its signature visual aesthetic.

The web-only release comes at a tense moment for the startup, dropping just days after Disney and Universal hit Midjourney with copyright infringement lawsuits. But inside the app, the focus is squarely on moving pixels.

How Midjourney V1 works

V1 is strictly an Image-to-Video tool for now. Instead of prompting raw text into video, users start with an image and click Animate.

The system gives you two paths: an automatic mode that generates motion prompts on its own, and a manual mode where you describe specific camera angles and character movements. You can also pick between high motion and low motion settings. Low motion keeps the camera steady for subtle movements, though it sometimes doesn't move at all. High motion creates dynamic action across the scene, at the risk of occasional artifacts.

Users can animate native Midjourney generations or drag in external images as starting frames. Every generation creates four separate 5-second video options. If you like one, you can extend it roughly 4 seconds at a time, up to four times, capping out around 20 seconds.

In terms of pricing, Midjourney charges roughly 8x the cost of a single image generation per video job, which works out to about one image's worth of cost per second of footage. CEO David Holz claims this is over 25 times cheaper than competing video models on the market, with an unlimited relax mode coming to Pro and higher tier subscribers soon.

The stepping stone to open worlds

V1 skips audio entirely and lacks direct text-to-video prompt capabilities, meaning it won't directly replace full-featured models like Veo 3 or Sora today.

However, Holz explained that V1 is simply a stepping stone toward a much grander goal: real-time open-world simulations. Building that requires four core components: visual imagery, motion through video models, spatial navigation through 3D models, and ultra-fast real-time rendering. Holz indicated that Midjourney plans to build and ship these individual pieces over the next year before integrating them into a unified interactive engine.

Why it matters

For existing Midjourney users, V1 removes painful workflow friction. Until now, creators had to generate stylized stills in Midjourney and jump over to Runway, Luma, or Kling to animate them. Now they get four motion variants in seconds without leaving the site, all while preserving Midjourney's signature cinematic mood.

Strategically, claiming a price point 25 times cheaper than rivals is an aggressive shot across the bow for the video generation market. High compute costs have historically kept casual users away from generative video; pricing a four-clip batch at just eight image generations makes experimentation practically painless.

Still, the launch carries clear limitations and risks. Without sound or native text-to-video, it remains an iterative creative toy rather than an end-to-end production suite. Looming lawsuits from Disney and Universal also cast a shadow over external uploads and character animation. But if you already have a paid subscription, V1 makes animating your existing catalog fast, remarkably cheap, and genuinely fun.