Pick a depth. Each prompt opens in your AI pre-loaded with the lesson. Click a row to preview the prompt.
Reading about prompts only gets you so far. At some point you have to pick one video tool, learn its quirks, and generate a clip you can show someone. The tools change monthly — what matters is that you pick one today, ship something, and build the habit. Runway and Kling are the current leaders; Luma, Pika, and Wan are catching up. All of them accept the same structured prompt you've been practicing.
Shipping a first clip matters more than optimizing it, because iteration feedback only arrives after you generate — reading about prompts can't replace watching a model respond to your specific language. The workflow below walks through tool signup, a copy-pasteable structured prompt, and two comparison generations that reveal model variance without spending extra credits. Thirty minutes from zero to a saved clip is an achievable first session.
1. Sign up for one tool (Runway free tier → 125 credits; Kling free daily credits; Luma's free plan works).
2. Paste this prompt verbatim:
Subject: a single candle flame in a dark room
Setting: an empty wooden table, dust motes in the air
Style: 16mm film, slight halation, warm
Camera: extreme close-up, static, locked-off
Lighting: the candle is the only light source
Motion: the flame flickers and dances for 4 seconds
3. Generate. Wait. Don't tweak yet.
4. Generate a SECOND clip with the SAME prompt. Compare — you'll see the model's variance.
5. Now tweak one line (e.g. "Camera: slow dolly in over 4 seconds"). Regenerate.
6. Download your favorite. That's your first shot.