Gemini Omni 1.2 Flash Pricing: 15s Clips at ~$0.07/s (720p)
Contains affiliate links; your price doesn't change · Disclosure · How we research
TL;DR
Gemini Omni 1.2 Flash is Google's video generation and multi-turn edit model, generally available on the Gemini API and Google Cloud as of November 5, 2026. Native clips go up to 15 seconds (was 10s on 1.1). Official paid-tier rates: input $1.50/1M tokens; text out $9/1M; video out $12/1M. At 5,792 tokens per second of 720p, that is about $0.07/s (~$0.70 for 10s, ~$1.04 for 15s). Free tier is not available. From December 3, 2026, gemini-omni-1.1-flash routes to 1.2 and bills at 1.2 rates. Consistency, speech, and 'production-ready' claims below are vendor-reported.
How this article was made: This article is research-based. GA date, 15-second length, ~30% price cut, and draft-then-upscale workflow come from Google's DeepMind blog and Google Cloud blog (both November 5, 2026). Token rates, tokens-per-second footnotes, Free-tier status, and the December 3, 2026 1.1→1.2 routing note come from Gemini Developer API pricing. Consistency, speech quality, and "production-ready" wording are vendor-reported, not our benchmarks. We did not run live billed generations. Checked November 6, 2026 (Taipei). Display name: CommX.
Gemini Omni 1.2 Flash is Google's video generation and multi-turn edit model, generally available on the Gemini API and Google Cloud as of November 5, 2026. Model ID: gemini-omni-1.2-flash.
The useful release-day facts are length and price: native clips up to 15 seconds, and 720p video billed at about $0.07 per second from Google's own token footnote.
Price table by resolution
Rates below are from the Gemini Developer API pricing page for gemini-omni-1.2-flash. Checked November 6, 2026 (Taipei). Currency is USD. Google / Gemini products are not in our affiliate list; links are plain official URLs.
Base token rates (Paid tier; Free tier Not available):
- Input: $1.50 / 1M tokens (text / image / video)
- Text output: $9.00 / 1M tokens
- Video output: $12.00 / 1M tokens
Google's official footnote: billing uses 5,792 tokens per second of 720p video. At $12 / 1M video tokens that is about $0.07/s. Other resolutions use the same footnote math.
| Resolution | Tokens / second (official) | Effective $/s | ~10s clip | ~15s clip |
|---|---|---|---|---|
| 360p | 1,931 | ~$0.023 | ~$0.23 | ~$0.35 |
| 720p | 5,792 | ~$0.07 | ~$0.70 | ~$1.04 |
| 1080p (with upsampling) | 8,688 | ~$0.104 | ~$1.04 | ~$1.56 |
| 4K (with upsampling) | 17,376 | ~$0.209 | ~$2.09 | ~$3.14 |
Figures above are output video tokens only, converted from Google's published footnote. Input tokens for prompts, reference images, or short motion clips add more. We are not inventing a "per finished ad" price beyond that math.
Gemini Developer API pricing (official)
1.1 vs 1.2, and the Dec 3 routing date
Google's blog says Omni 1.2 Flash is about 30% cheaper than Omni 1.1 and raises native length from 10s to 15s. The pricing page still lists 1.1 for comparison:
| Omni 1.1 Flash | Omni 1.2 Flash | |
|---|---|---|
| Model ID | gemini-omni-1.1-flash |
gemini-omni-1.2-flash |
| Native max length (blogs) | 10s | 15s (3–15s in 1s steps) |
| Input / 1M | $1.50 | $1.50 |
| Text out / 1M | $9.00 | $9.00 |
| Video out / 1M | $17.50 | $12.00 |
| ~720p effective $/s (official footnote) | ~$0.10 | ~$0.07 |
| Free tier | Not available | Not available |
Routing: from December 3, 2026, API requests to gemini-omni-1.1-flash automatically route to gemini-omni-1.2-flash and bill at 1.2 rates. Plan migrations before that date if you still pin the 1.1 ID.
Draft cheap, then upscale
Google's own workflow note matches the token table: draft at 360p or 720p, fix details with multi-turn edits instead of regenerating the whole clip, then upscale keepers to 1080p or 4K. That keeps most iterations on the cheaper rows of the table above.
Other vendor claims on the same pages (mark as such, not our tests):
- Stronger reference adherence and less visual drift across longer shots
- Multi-turn conversational edits (swap product, background, lighting, aspect ratio)
- More natural speech / lip-sync; Cloud blog says audio references are allowlisted
- SynthID watermark on outputs; Provisioned Throughput on Cloud; "production-ready" is Google's wording
Google DeepMind Omni 1.2 Flash blog (official)
What this page does not cover
We are not inventing per-clip retail prices beyond the official tokens-per-second footnote. We are not claiming Free-tier access or open audio-reference access for every account. We did not run OpenRouter as a substitute price source for per-second video.
Chinese / 知乎 mirror: skipped on purpose (CONTENT-RULES).
Related reading
- Images 2.6 / GPT-Image-2.6 Flare vs Sunburst (image API token table; different modality than Omni video)
- OpenMail (agent email infra if the video agent also needs an inbox)
- Zeabur shutdown migration (hosting for apps that call video APIs)
Sources (checked November 6, 2026, Taipei)
- https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/
- https://cloud.google.com/blog/products/ai-machine-learning/gemini-omni-1-2-flash-generally-available
- https://aistudio.google.com/docs/pricing
FAQ
What is Gemini Omni 1.2 Flash?
Gemini Omni 1.2 Flash is Google's video generation and editing model that supports native clips up to 15 seconds with multi-turn conversational edits. It became generally available on the Gemini API and Google Cloud on November 5, 2026. Model ID: gemini-omni-1.2-flash.
How much does Gemini Omni 1.2 Flash cost?
Per Google AI Studio pricing (checked November 6, 2026 Taipei): Free tier Not available. Paid: input $1.50 per 1M tokens (text/image/video); text output $9 per 1M; video output $12 per 1M. At 5,792 tokens per second of 720p, Google's footnote equates that to about $0.07 per second (~$0.70 for 10s, ~$1.04 for 15s).
How long can Omni 1.2 Flash videos be?
Official blog and Cloud post say native generate and edit up to 15 seconds, up from 10 seconds on Omni 1.1. Duration can be set in one-second increments from 3 to 15 seconds.
What happens to Gemini Omni 1.1 Flash?
Per the pricing page: effective December 3, 2026, API requests to gemini-omni-1.1-flash automatically route to gemini-omni-1.2-flash and are billed at 1.2 rates. Omni 1.1 video out was listed at $17.50/1M (~$0.10/s at 720p).
Is there a free tier for Omni 1.2 Flash?
No. The Gemini Developer API pricing table lists Free Tier as Not available for gemini-omni-1.2-flash. Audio references on Cloud are allowlisted; do not treat them as open to every account.
Key takeaways
- Gemini Omni 1.2 Flash is Google's video generation and multi-turn edit model, generally available on the Gemini API and Google Cloud as of November 5, 2026.
- Native clips go up to 15 seconds (was 10s on 1.1).
- Official paid-tier rates: input $1.50/1M tokens; text out $9/1M; video out $12/1M.
SourcesGemini Developer API pricing · Google DeepMind Omni 1.2 Flash blog (official) · cloud.google.com/blog/products
Last updated
Cite this pageGemini Omni 1.2 Flash Pricing: 15s Clips at ~$0.07/s (720p) — https://www.agentcomm.net/articles/gemini-omni-1-2-flash/