How to Prompt for AI Video: Runway Gen-3 & Sora Movements Demystified
Text-to-video generators require a totally different language than image generators. Learn the camera movements, kinetic verbs, and structure that yield smooth film scenes.
1. The Kinetic Language of Video Prompting
When you prompt a static image model, you describe a static state. But when prompting video systems like Runway Gen-3 Alpha, OpenAI Sora, or Luma Dream Machine, you are directing a sequence over time. You need to write in a kinetic language that defines both **subject movement** and **camera movement**.
If you don't define the camera movement, the model will often default to a lazy zoom or cause the subject to warp erratically. By taking control of the virtual lens, you establish stable frame pacing.
2. Camera Movements: Verbs that Work
Modern video models are trained on professional cinematic footage. Using real film director commands yields the highest consistency:
- Dolly-In / Dolly-Out: The camera physically moves closer to or further from the subject, changing background perspective lines.
- Tracking Shot / Trucking Shot: The camera moves horizontally alongside a moving subject, matching its speed.
- Pan and Tilt: The camera remains stationary but rotates horizontally (pan) or vertically (tilt).
- FPV Swooping Shot: For drone or high-speed kinetic movements, creating rapid depth shifts.
A low-angle tracking shot, moving alongside the muddy boots of a hiker walking through a rain-drenched pine forest. The camera is low to the ground, kicking up light splash reflections as they step. Soft morning fog in the background, cinematic shallow depth of field.
3. Subject Action & Pacing
Avoid ambiguous descriptions like "dancing" or "running". Instead, define the pace and transition of the action. Describe the speed using temporal terms:
- Use
"slow-motion","fluid slow pace", or"fast-paced dynamic motion"to dictate the temporal weight. - Describe actions with clear physical bounds: e.g.
"a drops of rain falling and hitting the leaf, causing it to bounce slightly"instead of just "rain on leaf".
A close-up slow-motion shot of a pouring stream of dark espresso coffee landing inside a white ceramic cup. The coffee creates small circular ripples, foaming at the top with hazelnut-colored crema. Steam rises gently, backlit by warm golden morning kitchen light.
4. Structuring a Cinematic Video Prompt
To keep your video generations stable, structure your prompt using this 3-tier sequence:
- Camera Direction: Set the camera movement first (e.g.
"A slow crane shot rising from..."). - Core Action: Describe the subject, action, and motion pacing.
- Environment & Lighting: Define the environment colors, weather, lighting angle, and lens quality.
"photographed on 35mm film, volumetric sunset rays, lens flares, film grain".
A slow dolly-out shot starting from a close-up of an old leather-bound journal lying open on a mahogany desk. Dust motes float in a warm beam of afternoon sun shining through a dusty window. The camera slowly reveals a cozy library background with tall bookshelves. Golden hour warm volumetric lighting.
Extract Video Prompts Automatically
Upload any video to our Prompt DNA page. The seek engine will analyze frame sequences and reverse engineer the prompts automatically.
Go to Prompt DNA