Seedance turns one image into a 7-shot flooded-forest short with a single prompt
umesh_ai · x · 2026-09-16
A user shares an AI video workflow: Seedance (via Runway) generates a short from one image plus a prompt, then Topaz Astra 2 upscales it.
Prompt highlights:
- Hard specs: 16:9, 4K, 24fps, exactly 360 frames, 15 seconds, seven shots with per-shot timings.
- Story: a lantern-carrying traveler enters a flooded forest; something immense approaches; instead of retreating, the traveler lowers the light, and an ancient forest guardian finally bows to the tiny flame — perceived danger becomes recognition.
- Each shot must add new visual information and advance the encounter; no repeated establishing views or decorative inserts.
More from Multimodal
- Two-step workflow: Midjourney stylized art then Gemini makes it photorealistic — michaelrabone · 2026-09-16
- Redditor uses MiniMax H3 to make a Y2K-style pop music video with one prompt — Kylie-Elena190 · 2026-09-16
- Dotey demos adding subtitles to video with AI tools — dotey · 2026-09-16
- Open-source video editor lets Claude drive itself via MCP to produce a promo video in 1 hour — _lolcat_ · 2026-09-16
- Creator can't stop churning out AI-generated 'slop microdramas' for fun — Kyrannio · 2026-09-16
- StepFun's StepAudio 3 Realtime reasons while speaking, hits 90.6 on MMSU — stepfun-ai · 2026-09-16