The short answer
Gemini Omni 1.2 Flash generates synchronized video and native audio from text, image, and video inputs[1]. The headline changes are practical: clips run up to 15 seconds (up from 10 in Omni 1.1), and you can now revise a clip through conversational editing — swap a product, change lighting or weather, or switch aspect ratio while leaving everything else intact[2].
For teams already on the API, this is a drop-in upgrade rather than a new product. Omni 1.1 stays available during a four-week transition, then existing integrations reroute to 1.2 automatically[1].
Longer clips and conversational editing
The developer docs support whole-second durations from 3 to 15 seconds, with 10 seconds as the default[3]. Output defaults to 720p in 16:9 landscape, with 9:16 portrait also available. Teams can draft at 360p or 720p, then upscale chosen results to 1080p or 4K[1].
Conversational editing preserves unmentioned elements across turns, so you don't regenerate the whole scene for a small change. A previous-interaction identifier keeps context between API requests, and each request can use up to six reference images — including up to three per character, for up to two characters[1].
Video extension works separately from native generation: it appends 3–10 seconds per turn (using the final 10 seconds as context) up to 30 seconds total. It cannot prepend footage or insert material mid-clip[1].
Pricing
On the paid Gemini API, Gemini Omni 1.2 Flash is $12 per million video-output tokens, metered at 5,792 tokens per second of 720p video[4]. That works out to roughly $0.70 for a 10-second clip and $1.04 for a 15-second clip[1]. Text, image, and video input costs $1.50 per million tokens, and the model is not available on the free tier[4].
The 720p price is about 30% lower than before — Google frames the cut around a changed production loop where targeted revisions replace full regenerations, and higher resolution is reserved for selected outputs[1].
Quality claims and limits
Google says 1.2 better preserves faces, garments, packaging typography, and brand colors across camera angles, with less visual drift and clearer lip-sync[1]. DeepMind's side-by-side evaluations give Omni 1.2 a 62% visual-consistency win rate against 38% for 1.1 in text-to-video, and a 63% multi-reference identity-consistency win rate in reference-to-video[5].
There are explicit technical limits to note: the API fully supports English only (other languages aren't yet evaluated), and voice editing, uploaded audio inputs, and audio references aren't supported in the current API[1]. Fast sports action, rapid script changes, and complex physics remain in development[1].
Availability
Developers can prototype in Google AI Studio and use the paid Gemini API. The broader rollout covers Flow for scene direction and multi-turn edits, longer clips in Google Vids at no extra cost, a Gemini app rollout to Google AI Ultra subscribers, and campaign creation in Google Ads[1]. Google Cloud's GA announcement adds console and API access, enterprise security, and Provisioned Throughput[6].
Every API-generated video carries an invisible SynthID watermark that can be detected programmatically for provenance[1]. Photoroom is an early production customer, using the model to power its Video Ad Maker and Product Animator[6].
Migration from Omni 1.1
Gemini Omni 1.1 Flash remains available during a four-week transition. On December 3, 2026, API requests to 1.1 automatically route to 1.2 without code changes and use the lower 1.2 rates — so existing integrations get a defined migration date without an immediate rewrite[1].
Bottom line
Verdict: Gemini Omni 1.2 Flash is a meaningful upgrade for anyone already using Omni 1.1 — longer clips, cheaper 720p output, and iterative editing that makes targeted revisions practical. If you need voice editing or non-English support, hold off; those remain on the roadmap[1].
Frequently asked questions
How long a video can Gemini Omni 1.2 Flash generate?
Native generation runs from 3 to 15 seconds (10 seconds default). Video extension can append more, up to 30 seconds total[1][3].
How much does Gemini Omni 1.2 Flash cost?
$12 per million video-output tokens, or about $0.70 for a 10-second 720p clip and $1.04 for 15 seconds. Input costs $1.50 per million tokens[4].
Do I need to change my code when Omni 1.1 is deprecated?
No. On December 3, 2026, API requests to 1.1 automatically route to 1.2 without code changes and use the lower 1.2 rates[1].
Sources
- Google Blog — "Introducing Gemini Omni 1.2 Flash" (Nov 5, 2026) — blog.google
- Google AI Studio — Omni documentation — aistudio.google.com
- Google AI Studio — Omni docs (duration and defaults) — aistudio.google.com
- Google AI for Developers — Gemini API pricing (Gemini Omni 1.2 Flash) — ai.google.dev
- Google DeepMind — Gemini Omni models — deepmind.google
- Google Cloud Blog — "Gemini Omni 1.2 Flash is now generally available" — cloud.google.com
- Unite.AI — "Google's Gemini Omni 1.2 Flash Reaches General Availability" — unite.ai