Gemini Omni 1.2 Flash Pricing: $0.07/s for 720p Video
Google cut Gemini Omni 1.2 Flash video generation pricing by 30% to $0.07 per second at 720p, with billing tied to output token consumption and automatic 1.1 Flash routing starting December 3, 2026.
Google cut Gemini Omni 1.2 Flash video generation pricing by 30% to $0.07 per second for 720p output, down from Gemini Omni 1.1. The update also extends maximum clip length to 15 seconds while tying API charges directly to output token consumption.12
How Gemini Omni 1.2 Flash token billing works
Developers are not billed a flat per-second fee independent of resolution. Instead, billing is based on total output token consumption under Standard pricing of $12.00 per 1 million tokens. At 720p, generation consumes 5,792 tokens per second, which produces the headline $0.07 per second rate.2
This token-pegged structure means resolution choices directly multiply API spend. Teams forecasting video generation API costs need to model both clip duration and target resolution rather than relying on a single per-second estimate. For teams comparing API spend against consumer plans, our AI subscription prices tracker shows Google AI Plus at $4.99 a month, Google AI Pro at $19.99 a month billed monthly ($16.67 a month billed annually), and Google AI Ultra at $99.99 a month, with Ultra 5x at $99.99 and Ultra 20x at $199.99 a month.23
Resolution costs from 360p to 4K
Output token rates scale across four supported resolutions, with 1080p and 4K relying on upsampling. Choosing 360p for draft previews cuts token consumption to 1,931 tokens per second, while 4K upsampling reaches 17,376 tokens per second.2
| Resolution | Tokens per second | Cost per second |
|---|---|---|
| 360p | 1,931 | $0.023 |
| 720p | 5,792 | $0.07 |
| 1080p (upsampled) | 8,688 | $0.104 |
| 4K (upsampled) | 17,376 | $0.209 |

December 3, 2026 endpoint migration
Effective December 3, 2026, all API requests sent to Gemini Omni 1.1 Flash will automatically route to Gemini Omni 1.2 Flash and be billed at 1.2 rates. Teams still calling the 1.1 endpoint do not need to rewrite model strings to capture the 30% price drop, but they should update Google Gemini API pricing forecasts before that date so billing alerts reflect the new token-per-second tiers.12
Clip limits and reference controls
Gemini Omni 1.2 Flash supports native video generation and editing for clips between 3 and 15 seconds, configurable in one-second increments. It also supports multi-turn conversational editing across full 15-second clips, allowing teams to refine ad variants without restarting generation from scratch.1
For tighter visual consistency in multimodal AI app architecture, the model accepts combinations of character, product, or style images, motion clips up to three seconds, and audio cues.1