Slop TVNewsLatest
News

Gemini Omni 1.2 Flash ships 15-second clips at $0.07 a second

Native clips now run 15 seconds, up from 10 on Gemini Omni 1.1 Flash, and the model edits conversationally instead of regenerating.

Illustration: Gemini Omni 1.2 Flash ships 15-second clips at $0.07 a second
Illustration: SLOP TV News. Logos are trademarks of their owners.

Key takeaways

  • Google rolled out Gemini Omni 1.2 Flash on November 5, 2026 and cut the price 30% from Gemini Omni 1.1 Flash to $0.07 per second of 720p video.
  • Gemini Omni 1.2 Flash generates native clips up to 15 seconds, up from 10 seconds on 1.1, with any whole-second duration from 3 to 15 seconds available in the developer documentation.
  • The developer documentation says Gemini Omni 1.1 Flash stays available for a four-week transition, and on December 3, 2026 API requests to 1.1 route automatically to 1.2 at the lower rates.
  • Google says Gemini Omni 1.2 Flash supports multi-turn conversational editing and up to six reference images per request, and every output carries an imperceptible SynthID watermark.

Google rolled out Gemini Omni 1.2 Flash on November 5, 2026 and made it generally available across the Gemini API and Gemini for business. The headline change is a price: Google cut the model's cost 30% from Gemini Omni 1.1 Flash to $0.07 per second of 720p video, a figure both launch posts state.

Native clip length rises to 15 seconds, up from 10 seconds on 1.1, and the developer documentation lets you set any whole-second duration from 3 to 15, with 10 seconds as the default. Drafts render at 360p or 720p and can be upscaled to 1080p or 4K. For a creator, a 15-second social spot or product explainer now fits in a single generation, and a fix costs a follow-up edit rather than a full regeneration.

That editing is the other change. Google says you can refine the same clip turn by turn, swapping a product or background, changing the lighting or weather, or switching aspect ratio, while the parts you did not mention stay put. The documentation allows up to six reference images per request (up to three images per character for up to two characters), and clips of 11 to 15 seconds take one reference video of up to three seconds. Audio references sit behind allowlisting, according to the Google Cloud post.

Google also describes stronger reference adherence, with faces, garments, packaging typography and brand colors holding across camera angles, plus more natural dialogue, clearer lip-sync, distinct multi-speaker conditioning and fewer unwanted jump cuts on continuous camera moves. Those descriptions come from Google's own launch posts. Google Cloud names Artlist, Krea, Photoroom, Picsart and Runway among companies using the model, with Photoroom CEO Matt Rouif saying 15-second video was a top user request.

Gemini Omni 1.1 Flash powered Google Vids' free 1080p tier in September, and it sat eighth on Artificial Analysis's text-to-video leaderboard at Elo 1116 as of October 2, 2026, per SLOP TV's earlier ranking.

On the paid Gemini API, Unite.AI reports metering at $12 per million video-output tokens, about $0.70 for a 10-second clip and $1.04 for a 15-second clip, and says the model is not on the free tier. Existing integrations have a migration date: the developer documentation says 1.1 stays available for a four-week transition, and on December 3, 2026 API requests to 1.1 route automatically to 1.2 at the lower rates. Outputs carry an imperceptible SynthID watermark.

Gemini Omni 1.2 Flash is available now in Google AI Studio and the paid Gemini API, and Google Vids generates 15-second clips at no additional cost with a Google account or Google Workspace subscription.

Sources

  1. blog.google - the launch post: GA across the Gemini API and Gemini for business, $0.07 per second for 720p video, the 30% cut, 15-second clips, 3-to-15-second durations, reference inputs, draft-and-upscale, SynthID, product availability
  2. cloud.google.com - GA on Google Cloud, the same $0.07 per second and 15-second clips, audio references behind allowlisting, customer list including Runway and Photoroom
  3. aistudio.google.com - the model ID gemini-omni-1.2-flash, whole-second durations from 3 to 15 with a 10-second default, resolution options, six reference images per request, and the December 3, 2026 transition from 1.1
  4. unite.ai - per-token metering at $12 per million video-output tokens, about $0.70 for a 10-second clip and $1.04 for a 15-second clip, and no free tier
  5. sloptvnews.com - track record: Gemini Omni 1.1 Flash powered Google Vids' free 1080p tier in September 2026
  6. sloptvnews.com - where Gemini Omni Flash 1.1 stood: eighth on Artificial Analysis's text-to-video leaderboard at Elo 1116 as of October 2, 2026