PROMPTIFYERPROMPTIFYERDirect AI video like a filmmaker. Fill the 8 cinematic blocks — intent, subject, frame, lens, camera move, performance, light, continuity — and get a directed prompt for Sora, Runway, Veo, Pika or Kling.
Pick your branch, fill the 8 blocks, generate a directed cinematic prompt.
Text-to-Video: the model starts from nothing, so describe subject, style, environment, motion and light.
See how a vague idea becomes a shootable cinematic prompt.
"A scared man runs through a dark street. Cinematic, realistic."
What's missing: shot size, lens choice, camera movement, performance direction, lighting logic, screen direction.
"Medium-wide, chest-height 35mm tracking shot. A soaked courier sprints left-to-right through a rain-slick alley, glancing back over his shoulder. Hard breathing, coat snapping. Neon signs reflect on wet asphalt; cool ambient glow. He brakes beside a steel door — hold on the locked handle."
The complete filmmaker framework behind every prompt this tool builds.
What should the viewer notice, feel or understand? Every choice below must serve this.
Who/what is in frame, exact action, direction of movement. Concrete verbs only.
Shot size and angle: close-up for emotion, wide for context, drone for scale.
Lens choice changes perspective: 24mm exaggerates space, 85mm compresses it intimately.
One motivated move per shot — push-in for pressure, tracking to follow, static for tension.
Expression, eye-line, body momentum, realistic behavior — not just "running".
Motivated light source and direction, mood, environment. Only relevant details.
Start state → end state: costume, props, screen direction stay consistent.
Use one motivated move per shot — never movement for its own sake.
| Move | Use it when… |
|---|---|
| Static | Observation, control, building tension |
| Push-in | To increase attention or pressure on the subject |
| Pull-back | To reveal context or isolation |
| Tracking | To follow a subject moving through space |
| Pan / Tilt | To reveal something or redirect attention |
| Handheld | Subjective urgency — use sparingly |
| Crane / Orbit | To show scale or a relationship |
| Rack focus | To shift attention between depth planes |
The full Promptifyer toolkit — everything free, no signup.
Common questions about cinematic video prompting.
An AI video prompt generator helps you write cinematic instructions for AI video tools like Sora, Runway, Veo, Pika and Kling. Filmmaker Mode below builds your prompt from 8 blocks — intent, subject, frame, optics, camera, performance, light and continuity — the same structure professional filmmakers use.
Text-to-video describes everything: subject, style, environment, motion and lighting, because you start from a blank page. Image-to-video focuses on motion only — camera movement, subject action and timing — since the image already defines the subject, style and composition.
The prompts work with OpenAI Sora, Runway, Google Veo, Pika, Kling and Luma. The cinematic language (shot size, lens, camera movement) is understood by all modern video generation models.
One shot = one beat: 5–10 seconds of action, a single camera move, and one main action. If you need a sequence, write separate prompts per shot in order — first X, then Y, finally Z.
It means each generation should contain exactly one main action and one camera move. Cramming multiple actions or camera setups into one prompt confuses the model and produces muddy results.
Yes — completely free, no signup, no limits. Filmmaker Mode runs 100% in your browser; nothing you type leaves your device.