Gemini Omni 1.2 Flash pricing: ~$0.07/s at 720p, ~$1.04 output for 15s
Gemini Omni 1.2 Flash pricing by the second: ~$0.07/s at 720p and ~$1.04 video output for 15s. Input billed separately. Includes 360p–4K table, 1.1 comparison, free tier has no video. China availability not verified.
bufferAbout the author4 min read
Gemini Omni 1.2 Flash bills video output by the second. At 720p that is about $0.07 per second, or about $1.04 of output for a 15-second clip. Input tokens for text, images, and video references are billed separately at $1.50 per 1M tokens. Do not treat $1.04 as the full cost of a generation.
Figures checked on November 6, 2026, Beijing time, UTC+8, against Google’s Gemini API pricing page, the Omni 1.2 blog post, and the Google AI post. If a draft number disagrees with the pricing page, the page wins. Whether the API or app features work from mainland China or Hong Kong is not verified here.
Bottom line
- Paid Gemini API access includes
gemini-omni-1.2-flash. The Free Tier column says Not available. That matches our Gemini API free tier limits note that image and video models stay off the free tier. - Effective output rates from the pricing footnote: about $0.023/s at 360p, $0.07/s at 720p, $0.104/s at 1080p, and $0.209/s at 4K. 720p uses 5,792 tokens per second at $12.00 per 1M video output tokens.
- A 15-second 720p clip costs about $1.04 in video output; 10 seconds about $0.70. Add input on top.
- Google says the price dropped about 30% versus Omni 1.1. 1.1 is about $0.10/s at 720p. From December 3, 2026, calls to
gemini-omni-1.1-flashroute to 1.2 and bill at 1.2 rates. - Gemini App and Google Vids subscription perks are separate from API pricing. The blog says Vids can generate 15-second clips at no additional cost with a Google account or Workspace subscription, and that 15-second videos in the Gemini app are rolling out to Google AI Ultra subscribers. Treat plan limits as Google states them.
How the pricing page becomes a per-second rate
Omni 1.2 Flash lists input at $1.50 per 1M tokens (text / image / video), text output at $9.00, and video output at $12.00. The footnote converts 5,792 tokens per second of 720p video into about $0.07/s, and lists the 360p, 1080p, and 4K token rates above.
Input is a separate line. A short text prompt is usually a tiny fraction of a cent. Reference images or video raise input. Vertex pricing mentions Omni charging 1,120 tokens per image in one note; the AI Studio pricing footnote for Omni does not pin that image token count next to the $1.50 input rate. Budget the output seconds first, then add measured prompt tokens.
Output cost table
Video output only. Rounded from the published effective per-second rates; 15s 720p uses the page’s ~$1.04 figure.
| Resolution | Per-second output | 5s | 10s | 15s |
|---|---|---|---|---|
| 360p | $0.023 | $0.12 | $0.23 | $0.35 |
| 720p | $0.07 | $0.35 | $0.70 | $1.04 |
| 1080p | $0.104 | $0.52 | $1.04 | $1.56 |
| 4K | $0.209 | $1.05 | $2.09 | $3.14 |
Worked examples
One 15s 720p clip: about $1.04 output. A 500-token text prompt adds about $0.00075 input. One reference image at 1,120 tokens (Vertex-style count; not restated on the AI Studio Omni footnote) adds about $0.0017. The bill is still dominated by output.
100 identical 15s 720p clips: about $104 of output, plus input for each prompt and reference.
Draft at 360p, finish higher: the blog suggests drafting at 360p or 720p, editing in conversation, then upsampling. Five seconds at 360p is about $0.12 of output; fifteen seconds at 1080p is about $1.56 of output.
Omni 1.1 vs 1.2
| Item | Omni 1.1 Flash | Omni 1.2 Flash |
|---|---|---|
| Native max length | 10 seconds | 15 seconds |
| 720p effective output | ~$0.10/s | ~$0.07/s |
| 10s 720p output | ~$1.00 | ~$0.70 |
| 15s 720p output | No native 15s then | ~$1.04 |
| Video output list price | $17.50 / 1M tokens | $12.00 / 1M tokens |
| From 2026-12-03 | Auto-routes to 1.2; billed as 1.2 | Current GA model |
Free tier, App, and Vids are different ledgers
- Gemini API free tier: Omni 1.2 is Not available. Paid tier required for API video. See Gemini API free tier limits.
- Google Vids: blog wording says 15-second clips at no additional cost with a Google account or Workspace subscription. That is an in-product allowance, not an API unit price.
- Gemini app: blog wording says 15-second videos are rolling out to Google AI Ultra subscribers. How many seconds other plans get, and where the rollout stands, follows Google’s statement and what your account shows. For plan context see Google AI Pro vs Ultra.
China availability
Whether Gemini Omni 1.2 Flash on the API, Gemini app, or Vids works from mainland China or Hong Kong is not verified in this article. We do not claim it works or fails without a test. Use Google’s official region lists for availability.
How to estimate your own bill
- Pick resolution and duration; use the table for output.
- Read prompt and candidate tokens in AI Studio or your logs; price input at $1.50 / 1M and video output at $12.00 / 1M.
- Multi-turn edits accumulate input and any new video output.
- For cross-model comparisons, use the LLM API pricing comparison.