CreateForge

Two photos, two distinct roles

Smalltown Boy AI Video Generator

Create a Smalltown Boy AI video with two separate pet or adult photos. Choose the room duet or people template, then use the included motion and audio to guide a scream, dance and final camera approach.

Create your version

Seedance 2.0

The first pet or adult, closest to the camera.

The second pet or adult, a separate character.

JPEG · PNG · WebP · ≤25 MB

Motion reference

Sign in

The included clip guides motion and camera, not an exact frame-by-frame copy.

Audio reference

Sign in

Audio guides generation; the original recording and exact lip sync are not guaranteed.

Advanced prompt
Settings14s · 720P · 9:16
14s
Aspect Ratio9:16
Resolution

Add the required reference content before generating.

Third-party reference clip, not a CreateForge generated result. Enable sound to hear the reference.

Prompts to try

Two-pet room duet

Apply in Pets mode with two pet photos.

Use image 1 only for the first pet, the foreground screamer; use image 2 only for the second pet, the dancing partner. Preserve each pet's species, face, ears, coat colors and markings separately. Follow video 1's room, framing and action progression: first a close foreground open-mouth cry while the partner dances behind; then both dance side by side; finally both move toward the lens. Keep each head on its own body, natural animal proportions and paws, no human faces, no extra animals, no identity swaps. Audio 1 guides the cry and musical rhythm. Generate scene audio without captions or logos. Maintain two distinct recognizable pets through the final approach.

Two-person garden duet

Apply in People mode with two adult photos.

Image 1 identifies the first adult, who performs the foreground scream; image 2 identifies the second adult, the dancing partner. Use video 1 as a motion and camera guide, replacing both reference animals with these two adults in their own clothing. No animal faces, tails or paws remain. Keep the garden fence setting and sequence: a close foreground cry, the partner dancing behind, both dancing side by side, then both approaching the lens. Preserve each adult's distinct face, age and skin tone. Natural human anatomy, separate bodies, no role swaps or extra people. Audio 1 guides the cry, rhythm and generated sound. No captions or logos.

  1. 01

    Assign two separate photos

    Photo 1 is the foreground screamer. Photo 2 is the dancing partner. Upload order never changes these roles.

  2. 02

    Choose pets or people

    Pets uses the room reference; People uses the garden reference with human replacement instructions. Switching replaces both included references and refreshes the quote.

  3. 03

    Review all three beats

    Check the scream, side-by-side dance and final approach, including both identities and the sound.

The Smalltown Boy AI scream-and-dance sequence

This page targets the two-character scream, dance and camera-lunge effect, not an unrelated dance in a cafe. The default Pets template uses two photos and the bright-room reference. People uses a different garden clip as its movement guide, with both characters changed into adults.

Each template sends two images, a silent fourteen-second motion clip and its extracted audio to Seedance 2.0. The sound is generated from a reference, not pasted from the recording; music, cries and timing may differ. Keep both photos distinct and inspect the ending where faces approach the camera.

Keep two characters from merging

Use one subject in each photo with a visible face and enough body detail. For pets, show ears and coat markings; for people, use clear portraits without masks. If roles merge, remove extra choreography and keep the first and second character instructions. Motion transfer and exact likeness are not guaranteed. Use only photos and audio you are permitted to use.

Credits and saved results

Sign in to upload references and review the current credit quote. Editing a prompt or changing a photo or setting clears the old quote. Generation starts only when you submit; opening this page or choosing a prompt does not spend generation credits.

Review the finished media before sharing. Successful results appear in Library with actual credit use. Result file URLs are public; browsing Library and task history still requires an account. References use temporary uploads, so a page refresh requires selecting photos again.

Common questions

This tool requires two separate photos so the roles stay explicit. Photo 1 screams first; Photo 2 is the partner. A missing or unfinished photo blocks quoting.

The two photo positions stay in memory. The included motion and audio change, and a new quote is required. Replace photos if the new mode needs different subjects.

Yes. The server quotes the selected model, settings and references before submission. This page does not promise a free generation or copy another website's credit prices.

References guide appearance, but likeness, hands and motion can vary. Use a clear photo and inspect the result. A completed task is not a guarantee that every creative detail is correct.