Reverse-engineer any viral AI video's prompt with ChatGPT in seconds
TawohAwa · x · 2026-10-19
Creator TawohAwa shares a full workflow for reverse-engineering any viral AI video instead of begging for or buying prompts:
- Download the video from any social platform with a free downloader tool.
- Upload it to Claude, ChatGPT or Gemini with the prompt: "Watch this video frame by frame. Then create a prompt for me specifically for Seedance 2.5 to recreate this video perfectly." Optionally attach a photo of yourself to swap in as the main character.
- Paste the generated prompt into a video generation platform — the author uses Syllaby, which aggregates models like Veo, Seedance, Kling and Grok Imagine (affiliate link).
The core trick is using multimodal LLMs to extract a replication prompt from any video, then rebuilding it with a video model like Seedance 2.5.
More from Multimodal
- MiniMax-H3 hailed as favorite open release of the year, runs fully local — AdinaYakup · 2026-10-23
- Michal Malewicz experiments with depth maps in his creative workflow — michalmalewicz · 2026-10-23
- One-Shot Tunable Style Transfer on Flux3 Via Reference Latent + Node — Winter_unmuted · 2026-10-23
- PhysVidBench accepted at EMNLP 2026: do video gen models grasp physical commonsense? — aykuterdemml · 2026-10-23
- EMNLP 2026 paper teaches multimodal LLMs to get cartoons via incongruity-resolution supervision — aykuterdemml · 2026-10-23
- A gorgeous Midjourney style: sref code 820221253 for dreamy lavender watercolor looks — umesh_ai · 2026-10-23