Gemini Omni 1.2 Flash: 15-Second AI Video at $0.07 a Second

Gemini Omni 1.2 Flash: 15-Second AI Video at $0.07 a Second

Published
AuthorDevansh

On November 5, 2026, Google DeepMind made Gemini Omni 1.2 Flash generally available across the Gemini API and Gemini for business, with Google Cloud announcing the same model for enterprise customers the next day. It is the fast, lower-cost video generation and editing model in the Gemini family, and this update is a cost and control upgrade rather than a new capability: clips run up to 15 seconds instead of 10, the price fell about 30 percent to roughly $0.07 per second of 720p video, and editing is now multi-turn, so you can keep refining a clip in conversation instead of starting over. The API model ID is gemini-omni-1.2-flash.

What changed from Omni 1.1

Four things. Clip length: a generation can now run 3 to 15 seconds, set in one-second steps, up from a 10-second cap, and video input for editing runs up to 15 seconds with a 10-second lookback for extension. Price: video output dropped from $17.50 to $12.00 per million tokens, which the pricing page puts at about $0.07 per second at 720p. Editing: multi-turn conversational editing through the Interactions API, with stronger reference following so a character or product stays consistent across edits. Availability: the model is production-ready on the Gemini API paid tier and Google Cloud with Provisioned Throughput, and in the Gemini app 15-second videos are rolling out to Google AI Ultra subscribers. Outputs still carry an imperceptible SynthID watermark so a clip can be checked as AI-made.

What it costs in practice

Billing is token-based, at a fixed 5,792 tokens per second of 720p video. Input costs $1.50 per million tokens and video output $12.00 per million, which works out to about $0.70 for a 10-second 720p clip and about $1.04 for a full 15-second clip. A 360p draft runs about $0.023 per second, so a 15-second draft is about $0.35. Two limits to plan around: there is no free tier on the Gemini API for this model, and it sits on the paid tier only. For a team generating video at volume, the budgeting rule is simple: a hundred 15-second 720p clips is about $104, and the per-clip cost is low enough that the real budget line is review time, not tokens.

The December 3 migration to plan for

From December 3, 2026, every request to the old gemini-omni-1.1-flash model ID is routed automatically to 1.2 and billed at 1.2 rates. If you pinned 1.1 anywhere, check those call sites now rather than discovering the change in an invoice: the output will be 1.2 quality at 1.2 prices, which is a good deal, but a silent model swap is still a change worth knowing about. The same discipline applies to any model you build on: pin the version, watch the deprecation dates, and re-run your evals when the provider moves you. Our Gemini 4 Carbon explainer covers the flagship model in the same family and its own December 1 price change, and our n8n AI agent guide covers automating a content pipeline around a model like this without letting it run unsupervised. If you want the model choice, the spend alerts, and the monitoring around an AI pipeline checked by someone who runs infrastructure for a living, book a free infra audit with DevOpsBY.

Disclaimer: This post is educational only, not financial advice. Figures are as reported by the sources on 2026-11-17 and may change - always confirm version numbers, requirements, and prices in the official docs before acting on them.

Related Articles

DevOps & Infrastructure Services

Need hands-on help? We build and manage the infrastructure behind the concepts covered in this article.

DevOpsBy Assistant

How can we help?

Select a topic to get started: