Midjourney has just introduced its first AI video model, Midjourney V1, which transforms static images—user-uploaded or AI-generated—into short, animated clips. The feature is now available via Midjourney’s web interface and Discord channels, catering to creative professionals and enthusiasts.
How Midjourney’s Image-to-Video Feature Works
Once an image is generated or uploaded, users can click the new “Animate” button to begin the transformation process. The Midjourney V1 model produces four unique five-second video clips per image.
Two animation presets are offered:
-
Low motion – subtle movements like panning or tilting.
-
High motion – dynamic motions including rotations or zooms, with increased risk of visual artifacts.
Users can also use text prompts to guide the animation and extend video clips by four seconds, up to a total duration of approximately 20–21 seconds per video.
Pricing Model and Subscription Plans
Midjourney V1 model is included in all existing subscription tiers, but video generation consumes approximately eight times more GPU credits than a single image. The following table outlines the available plans and their benefits:
| Plan | Monthly Cost | Annual Cost (20% Discount) | Fast GPU Time / Month | Relaxed GPU Time | Stealth Mode | Ideal For |
|---|---|---|---|---|---|---|
| Basic | $10 | $96 ($8/month) | 3.3 hours | None | No | Casual users testing AI video features |
| Standard | $30 | $288 ($24/month) | 15 hours | Unlimited | No | Regular creators balancing speed/cost |
| Pro | $60 | $576 ($48/month) | 30 hours | Unlimited | Yes | Professionals needing privacy and speed |
| Mega | $120 | $1,152 ($96/month) | 60 hours | Unlimited | Yes | High-volume users and agencies |
All users can purchase additional Fast GPU time for $4/hour. Video generation is available under both Fast and Relaxed modes, with rendering times varying accordingly.
Video Quality and Artistic Focus
Midjourney’s V1 does not attempt full photorealism. Instead, it leans into the platform’s signature style—dreamlike, abstract, and expressive. Early users note that low motion clips retain visual coherence, while high motion clips are more experimental but sometimes show flickering or distortion.
Though visually captivating, Midjourney’s V1 is seen as less realistic than competitors such as OpenAI’s Sora or Google’s Veo. However, the creative flexibility and aesthetic flair have been praised by artists and animators.
Competitive Positioning and Strategic Direction
With the launch of V1, Midjourney enters direct competition with AI video tools like:
-
OpenAI’s Sora
-
Google’s Veo
-
Adobe’s Firefly Video
-
Runway’s Gen-4
-
Luma’s Dream Machine
Rather than focusing on cinematic realism, Midjourney aims for interactive and expressive content creation. According to founder David Holz, this release is a stepping stone toward more advanced capabilities, including real-time rendering, 3D simulations, and eventually fully explorable AI-generated worlds.
Legal and Technical Complications
The Midjourney V1 release comes amid an ongoing copyright lawsuit filed by major studios including Disney and Universal. The case alleges that Midjourney’s training data includes copyrighted material. While Midjourney has not commented extensively on the litigation, it has emphasized its commitment to creative, non-commercial use cases.
Legal outcomes could impact future model development, data sourcing, or usage limitations, though the V1 rollout continues uninterrupted.
What Can We Expect Midjourney Video AI to Improve Next?
Key developments to watch in the coming weeks:
-
Updates to the cost structure and GPU consumption for V1.
-
Improvements to video stability and reduction of artifacts.
-
Additional controls for motion and style.
-
Expansion into 3D rendering and possibly audio integration.
-
Legal decisions that may affect training data and content policy.
