Antigravity reverse-engineers video prompts, wires Omni model to auto-remake viral clips
mazzaTalk · x · 2026-11-20
leafsanren shares a workflow combining Gemini 4 + Antigravity to fully automate viral-video replication. Antigravity can't generate video itself, but it can deconstruct a reference video's multi-shot cuts, framing, and lighting into a prompt, then call the gemini-omni-1.2-flash video API (16:9, 360p, 8s) to re-render the clip with only the subject swapped for a custom character.
The post includes the full call structure: input order is silent reference video → character sheet → key-frame image → targeted prompt demanding a 1:1 remake of the original except the white cat. A longer write-up covers running the whole auto-production pipeline on Google Pro/Ultra subscription benefits.
More from coding & agent
- CapCut plugin connects ChatGPT to auto-cut gameplay highlights, then human picks the payoff — iamfakhrealam · 2026-11-21
- Dev wires a Runway-generated avatar into Nous Research's Hermes Agent, bashing boring agent UIs — alexcovo_eth · 2026-11-21
- Lean compiles, but the natural-language proof is still wrong — jasondeanlee · 2026-11-21
- DeepMind demos ski simulator built by Gemini 4 Neon from real Mont Blanc terrain — GoogleDeepMind · 2026-11-21
- pstack v0.15.46 trims skills, adds Gemini as panel reviewer; pinning question raised — petrusenko_max · 2026-11-21
- Hamel Husain on evals: a good process beats good tools — HamelHusain · 2026-11-21