Midjourney Launches V1 AI Video Model, Generating 5‑Second Clips from Images

Midjourney, an AI picture‑generating company, announced on 18 June 2025 the release of its highly anticipated AI‑video generation model, V1.

V1 is an image‑to‑video model that allows users to upload an image or select one generated by Midjourney’s other models; V1 creates four five‑second videos based on it. Like Midjourney’s picture models, V1 is only available through Discord and will be exclusively accessible on the web at launch.

The release of V1 positions Midjourney in direct competition with existing AI‑video‑generating models from OpenAI, Runway, Adobe and Google.

Unlike its competitors, which focus on developing controlled AI‑video models for commercial applications, “Midjourney’s approach remains unique.”

V1 Model

The V1 model offers various configurable options to give users greater control over video outputs. Users can choose an auto setting to make an image move randomly or a manual mode to specify the desired animation. The level of camera and subject movement can be adjusted by selecting “low motion” or “high motion” in the settings.

While V1 videos are initially five seconds long, users can extend them by up to four seconds at a time, enabling videos to reach a maximum length of 21 seconds.

Early examples of V1’s videos appear more otherworldly than hyperrealistic, akin to Midjourney’s AI picture models. The initial response to V1 has been positive, but it remains unclear how it compares to other prominent AI‑video models that have been available for months or even years.

V1 Price Point

Midjourney has announced that its V1 video‑generation model will cost eight times more than standard image generation, meaning users will deplete their monthly generation quotas much faster when creating videos.

The most cost‑effective way to try V1 at launch is through Midjourney’s $ 10‑per‑month Basic plan. The $ 60‑per‑month Pro plan and $ 120‑per‑month Mega plan offer unlimited video production in the slower “Relax” mode, which queues jobs and may take longer to process. Midjourney intends to review its video‑model pricing within the next month.

Midjourney CEO David Holz revealed in a blog post that their AI‑video model is merely a stepping stone toward a broader objective: developing AI models capable of real‑time open‑world simulations.

Following the V1 release, Midjourney plans to develop AI models for 3D rendering and real‑time applications.