Betrayal and a Slap on a Subway Platform
Drama & story · Seedance 2.5 · @AIwithSynthia ·
A 30-second text-to-video prompt that generates a live-action melodrama on a modern subway platform: a young woman with coffee and phone catches her boyfriend kissing someone else, drops both, confronts and slaps him, then walks away in tears. All three characters are written out in words, following the clip.
Original video: @AIwithSynthia · View post ↗
Cost on Makernice
$7.776
Seedance 2.5 · text-to-video · 720p · 30 s × $0.2592/s
- 480p$3.45888
- 720p$7.776
- 1080p$19.2456
- 4k$58.65
- Failed, timed-out or canceled-while-queued runs release the hold in full. How it works
- Quote first: free, locked for 12 minutes, charged exactly the quoted amount. API
- A finished clip blocked by output moderation is charged; manual review is available.
Prices as of November 3, 2026 (price list v12).
Prompt
Run it with an AI agent
- Create an API key with its own spend limit.
- Copy the agent brief and paste it into Claude Code, Codex or another coding agent.
- The agent quotes first and waits for your OK, then creates with an Idempotency-Key and polls until the video is ready.
Why it's adapted
The creator's prompt relies on uploaded photos to fix the three faces; this adaptation removes that dependence and writes every outfit and hairstyle in words, so it runs as plain text-to-video. The finished clip mostly follows the written story but changes the order and a few details: the opening shot already shows the couple kissing behind her as she walks past, the moment the cup and phone fall is never shown and only a floor insert of the aftermath appears, the other woman stays pressed against the pillar instead of stepping back, the boyfriend does not visibly follow or call after her, and in the last close-up her gaze only drifts near the lens. The adaptation follows the clip, keeps the creator's sound plan and most of the dialogue, folding two lines into one, and renders as one 30-second Seedance 2.5 text-to-video at 16:9 and 720p; the prompt asked for 1080p and the published file is 1280×720.
Original prompt
@AIwithSynthia · 3,795 characters
Shot breakdown
- Subject
- Three original characters told apart by clothes: the heroine in a light-blue cropped top, high-waisted jeans and white sneakers with small silver jewellery; the boyfriend in a charcoal jacket, black tee and black trousers; and the other woman with wavy light-brown hair, a tie-neck cream blouse and a short black skirt. The station is a clean, modern platform lined with glass screen doors and tiled pillars.
- Action
- Discovery, aftermath, confrontation, exit: she walks past the kissing couple unaware, sees them, and the spilled coffee and dropped phone are shown after the fact; she marches over, argues, slaps him when he grabs for her arm, wipes her tears, retrieves her phone from the puddle and walks off down the platform as a train approaches.
- Camera
- Six shots, most of them long: a backward-tracking walk that pushes into a close-up as she looks up, a floor-level insert of the spill, an extreme close-up of the first tear, then one unbroken handheld take of about 14 seconds that moves from behind her into a three-shot for the argument and slap, a low shot by the puddle that rises into a follow from behind, and a closing close-up in a soft-focus crowd.
- Style
- Glossy live-action drama under cool fluorescent light, with shallow focus on faces, natural crying and a handheld feel that stays readable. The commuters react but stay in the background, so the emotion sits on the three leads.
- Sound
- Station ambience and distant announcements, the clatter of the cup and phone, three short exchanges of trembling dialogue, a slap that echoes off the tiles, uneven breathing, train brakes and an optional soft piano as she walks away.
Recommended settings
- Model
- Seedance 2.5
- Duration
- 30 s
- Aspect ratio
- 16:9
- Resolution
- 720p
Make it yours
- If the three characters start to blend, lean harder on the clothing contrast in the character lines (light blue versus charcoal versus cream) rather than adding facial detail; inside the long handheld take, colour is what keeps them apart.
- The slap reads best as a short, clearly separated beat, as written around 12s; if it gets swallowed by the argument, trim the dialogue before it rather than stretching the slap.
- To make it less intense, swap the slap for her pulling her arm away and stepping back, and remove the slap echo from the sound line so the audio still matches.
FAQ
Do I need photos of the actors to keep the three characters consistent?
No. This version describes each person's hair, clothes and accessories in full, so it works in text-to-video mode on its own. If you do want to supply your own photos, Seedance 2.5 has a reference-to-video (r2v) mode that accepts up to 16 images; trim the written character lines there so the words and the photos agree.
Which resolution should I choose for the tear close-ups?
Seedance 2.5 renders at 480p, 720p, 1080p or 4K, and this page uses the 720p default. Tears and wet eyes hold up better at 1080p or 4K, so test a short cut at 720p first, then render the full 30 seconds at the higher setting.
Can I split this into two shorter scenes?
Yes. Each Seedance 2.5 run can be 4 to 30 seconds, so you could generate the discovery and slap (0-13s) as one clip and the aftermath and walk-away (13-30s) as another, renumbering the timecodes from zero in the second prompt and keeping all three character lines word for word in both.
Source and copyright
The video and original prompt come from a public post on X by @AIwithSynthia. Copyright belongs to the original creator; this page is for reference only. View on X
For any copyright concern, contact support@makernice.com and we will handle it promptly.
More Drama & story prompts
Drama & storyTowel Dash to the Door for a Missed Delivery
A 15-second photoreal comedy prompt that generates a man showering when a delivery alert lights up his phone, a peephole glimpse of the waiting driver, a barefoot sprint down the hallway in a towel, and a front door opened just as the driver walks off. It is written from the finished clip, cut by cut.
Seedance 2.5
Windy Laundry Chase in an Apartment Yard
A short comedic prompt in labeled blocks (character, location, camera, sequence, audio): a gust blows laundry off a line and she keeps chasing it, finishing with a look to the camera.
Seedance 2.5
Favorite Song Starts Playing On The Sidewalk
A short comedic POV prompt: a man hears his favorite song from a café, slows down and vibes while pretending to keep walking. About half the text is a list of realism and physics terms.
Seedance 2.5
Glass Sphere to Typographic Match Cuts
A concept-driven title-sequence prompt built on one continuity rule: each shot must inherit an object, shape or letter from the previous one. It gives a worked example chain and then asks the model to continue it.
Seedance 2.5