Seedance 2.5

Seedance 2.5 — up to 30s in one shot, native audio, wide multimodal references. 480p/720p/1080p.

Seedance 2.5 examples

Credits

Use Seedance 2.5 in Nodaro

Add one of these nodes to your canvas, open its settings, and choose Seedance 2.5 as the model:

• Generate Video
• Generate Video Pro
• Edit Video Pro
• Lip Sync

Run Seedance 2.5 from code

Each call starts a job and returns its jobId. Poll GET /v1/jobs/:id/status, or let the SDK wait for the result. A run costs the credits on this page. API reference

With Nodaro connected to Claude, ChatGPT or Cursor, ask in plain words, or have the assistant call the tool: “Use Seedance 2.5 to make: A paper boat drifts down a rain-soaked street at dusk, slow dolly in”

curl -s -X POST https://app.nodaro.ai/v1/text-to-video \  -H "Authorization: Bearer $NODARO_API_KEY" \  -H "Content-Type: application/json" \  -d '{        "prompt": "A paper boat drifts down a rain-soaked street at dusk, slow dolly in",        "provider": "seedance-2-5",        "duration": 8      }'

import { createClient, StaticTokenAuth } from "@nodaro/sdk"const client = createClient({  baseUrl: "https://app.nodaro.ai",  auth: new StaticTokenAuth(process.env.NODARO_API_KEY!),})const output = await client.nodes.runAndWait("text-to-video", {  prompt: "A paper boat drifts down a rain-soaked street at dusk, slow dolly in",  provider: "seedance-2-5",  duration: 8,})

{  "tool": "generate_video",  "arguments": {    "prompt": "A paper boat drifts down a rain-soaked street at dusk, slow dolly in",    "model": "seedance-2-5",    "duration": 8  }}

Prompt tips

• Storyboard as 'Shot 1: … Shot 2: …'. The 2.0 SKUs ignore timestamps and '(0-3s)' shots are officially unstable there; seedance-2-5 honours integer-second timestamps ('0-3s: …', 'at the 5-second mark').
• One camera movement per shot; describe actions per body part with degree ('slowly raises a hand'); express emotion as physical detail, never abstract words.
• Native multi-track audio — cue it inline: （background music）, <sound effects>, and quoted dialogue.
• References go by ordinal (@Image 1, Video 2) in attachment order; earlier = higher priority. Identity = ONE headshot + ONE full-body (multi-view sheets drift on 2.0; 2.5 accepts them). 4-5 assets beats maxing 9/3/3.
• No negative-prompt parameter — put constraints in the prompt: 'keep it subtitle-free, do not generate a watermark, do not generate a logo'.
• seedance-2-5 only: one shot runs to 30s (2.0 stops at 15s) — storyboard a whole beat, not a stitch; timestamps work (gap-free, 1s units, don't overpack); refs 30/10/10 but 4-5 assets still gives the best identity.
• Auto-path formula: Subject → Action → Environment → Camera → Style → Constraints in 60-100 words; ONE camera instruction (chain with 'then'); separate camera motion from subject motion; always add one lighting phrase.

Similar models