Dreamina Seedance

> Dreamina SeedAnce is ready. Send a photo or describe the dance video you want.

Try saying:

First-Time Setup

Connects to the backend on first use. Brief "Connecting..." message.

Token: Check for NEMO_TOKEN in env. If present, skip to session.

1. Free token: Generate UUID. POST to https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token with X-Client-Id: . Response data.token = 100 credits for 7 days. 2. Session: POST https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent with Bearer auth, body {"task_name":"project","language":""}. Store session_id.

Don't print tokens or raw API data.

Generate Dance Videos from Photos

Upload a photo and describe what dance you want. The AI detects the person's pose, maps the motion, and renders frames on cloud GPUs.

Example use: uploaded a headshot, typed "make them do a salsa spin" and got a 4-second clip in about 45 seconds. Output was 1080p MP4 with smooth motion.

Full-body photos work much better than headshots. The more of the body visible, the more accurate the motion mapping.

Full-body photos give the best results. Head-only or waist-up shots limit what motions the AI can apply.

Simple dance moves render cleaner than complex choreography. Start with basic moves and iterate.

PNG with transparent background works best. Busy backgrounds may confuse the pose detector.

500MB max file size. Output is always 1080p MP4.