Prompting motion for video control
Midjourney’s video generation capabilities are no longer just about visual style—they now respond directly to your descriptions of motion, camera behavior, and mood. In this section, you’ll learn how to shape animation using prompt language and supported parameters. You’ll also see examples where a still image combined with a motion-aware prompt produces expressive results.
Prompting to control animation
In Midjourney, a prompt does more than set the look of a scene—it determines how that scene moves. This is what makes video prompting distinct from still image generation. When you write a prompt for a video, the AI interprets it through a cinematic lens: it considers what should move, how quickly, what kind of camera behavior to simulate, and what emotion or tone the scene should carry.
Example:
a glowing jellyfish drifting through deep ocean currents, cinematic lighting, slow motion
This prompt does more than describe a scene. The model will attempt to animate the drifting, simulate underwater parallax, and apply smooth, low-frame motion. Words like “drifting,” “cinematic,” and “slow motion” all contribute to how the video flows.
To guide animation intentionally, try including elements like:
- Subject movement (e.g., flying, pulsing, walking)
- Camera behavior (e.g., zoom in, pan right, orbit)
- Tempo (e.g., slow motion, fast-paced, rhythmic)
- Mood and tone (e.g., melancholic, dreamy, chaotic)
- Visual treatment (e.g., watercolor, cinematic, cel-shaded)
You don’t need to follow a strict format—just describe the scene as if you’re directing it. For instance:
a lone violinist on a foggy street, warm backlight, slow tracking shot, melancholic atmosphere
This type of prompt not only conveys visual content but also how the camera should move and how the scene should feel.
You can take control even further by combining a reference image with your prompt. For example, dragging in a character image and prompting:
slow camera orbit, glowing embers in the background, soft dusk lighting
This type of prompt will anchor the composition to your image while animating around it. This technique is helpful when you want to:
- Animate static portraits or product renders
- Add life to hand-drawn scenes
- Preserve character identity while adding cinematic flow
Using soft instructions like:
gentle zoom in, ambient lighting shift
results in minimal transformation and keeps the image intact.
Stronger instructions like:
explodes into stardust, camera spins upward
push the animation toward dramatic motion and visual change.
Using video-specific parameters
In addition to natural language prompts, Midjourney currently supports two officially documented parameters for video generation:
--motion <low|high>
--motion low(default) produces gentle, stable animations, such as soft drifts or ambient movement.--motion highincreases the intensity and range of animation, suitable for dynamic, energetic scenes. However, higher motion can sometimes introduce visual inconsistencies or glitches.
--raw
The --raw flag disables some of Midjourney’s default stylistic interpretation, giving your prompt more direct control over how the video behaves. This is especially helpful when you want to preserve the original structure of a reference image or avoid abstract visual embellishments.
Case studies: Prompt and motion in action
Let’s look at three real-world examples of how a still image, when paired with a motion-aware prompt, results in a unique animated sequence.
Case 1: Neon anime girl with musical energy
Image: A colorful anime girl laughing with headphones, glowing under vivid lighting.
Prompt:
neon-lit anime girl laughing joyfully, wearing headphones, camera slowly rotates around her, vivid pink and yellow glow, hair fluttering with rhythm, lighting pulses with music beat --motion high
Watch the video here to see it in action.
Blending One Character into Different Worlds
The result is a lively, music-driven animation with smooth camera rotation, pulsing lights, and expressive hair movement that brings the scene to life.
Case 2: Cinematic kitten in a quiet street
Image: A stylized 3D kitten in a hoodie, standing on a rainy street and looking up.
Prompt:
adorable animated kitten on a rainy city street at night, wearing a hoodie, soft camera push-in, gentle raindrops falling, warm bokeh lights, eyes glistening with curiosity --motion low --raw
Take a look at the video here. The video on this page includes the extension scene that will be explained in the next section.
Testing Style and Character Reference Impact
You’ll notice how the warm lighting and slow movement come together to create a quiet, cinematic moment filled with emotion.
Case 3: Cool cat overlooking a sunny harbor
Image: A cartoon cat with sunglasses, leaning on a balcony with ships and clouds behind.
Prompt:
cool cartoon cat in sunglasses overlooking a seaside harbor, gentle breeze swaying ship flags, slow zoom out, drifting clouds, birds flying in the distance, relaxed summer vibe --motion low --raw
Watch the video here to see the full effect.
Putting the Same Cat into Different Scenes
You’ll notice how the drifting clouds, swaying flags, and smooth zoom-out combine to capture a relaxed, coastal vibe.
Prompting for video control in Midjourney is more than writing visual descriptions—it's about directing motion, mood, and camera flow using words. By combining clear prompt language with optional parameters like --motion and --raw, you gain deeper creative control over how your animations unfold. Whether you're crafting calm ambiance or dynamic action, thoughtful prompting gives you the tools to shape each scene with intention.