SSyncDev

Gemini Omni 1.2 Flash

SyncDev model directory · Profile updated Nov 17, 2026

Compare with another model

Gemini Omni 1.2 Flash is Google's conversational video generation and editing model, making 3-15 second clips with native audio from 360p to 4K.

Key specifications

Specification score
44.4
Context window
131.07K
Max output
65.54K
Input / 1M tokens
From $1.50
Output / 1M tokens
From $12.00
License
Proprietary

The specification score summarizes recorded features and limits. It is not a benchmark result.

Pricing conditions: Indicative per 1M tokens: input $1.50 (text, image or video); video output $12; text output $9. Video billed at 5,792 tokens per second of 720p (about $0.07/s). No free tier. Confirm in your Google billing console.

Inputs and outputs

Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.

Accepts

  • Text input
  • Image input
  • Video input

Produces

  • Video output
  • Audio output
  • Text output

Capabilities

VisionAudio

What is Gemini Omni 1.2 Flash?

Think of how a director works: shoot a take, watch it, say "again, but slower, and keep the red coat". Gemini Omni 1.2 Flash brings that loop to the API. It generates video with native audio from text, images or existing video, then lets you refine the result turn by turn through the Interactions API instead of starting each prompt from scratch.

It reached general availability on 5 November 2026 and is now Google's named replacement for its earlier preview video models.

Specification

Field Value
API ID gemini-omni-1.2-flash (stable)
Released 5 November 2026
Input Text, image, video
Output Video with native audio, plus text
Input token limit 131,072
Output token limit 65,536
Clip length 3 to 15 seconds
Extension 3 to 10 seconds per turn, up to 30 seconds total
Resolution 360p, 720p, 1080p, 4K
Frame rate 24 fps
Reference images Up to 6

Video input can run up to 15 seconds for editing, with up to 10 seconds of lookback for extension. First- and last-frame interpolation lets you pin where a shot begins and ends.

Consistency across shots

The six-reference-image limit is the feature that matters most for anyone making more than one clip. Keeping a character's face, a product's shape or a mascot's colours stable across shots has been the weak spot of generated video. Google says 1.2 improves character and subject consistency across those references. Test it with your own assets, because consistency claims are easy to make and hard to measure.

What a clip costs

Pricing is per token. Input of any type is $1.50 per million. Video output is $12 per million, 30% lower than 1.1. Text output is $9 per million. Video is counted at 5,792 tokens per second of 720p footage, which Google converts into rough per-second prices:

  • 360p: about $0.023 a second
  • 720p: about $0.07 a second, so roughly $0.70 for ten seconds and $1.04 for fifteen
  • 1080p with upsampling: about $0.104 a second
  • 4K with upsampling: about $0.209 a second

There is no free tier. Some gateway listings show text output at $9 and others at $12; Google itself prices text at $9 and video at $12, so a $12 text rate probably conflates the two. Check which rate your provider applies. All figures are indicative.

Paid-tier rate limits doubled compared with 1.1.

Migration deadlines

Three preview video models, Veo 3.1 Preview, Veo 3.1 Fast Preview and Veo 3.1 Lite Preview, were deprecated with a 22 October 2026 shutdown and point here. So does Gemini Omni Flash Preview, shut down on the same date. Anyone on gemini-omni-1.1-flash has less to do: requests to that ID route to 1.2 automatically from 3 December 2026. If your pipeline is sensitive to output changes, pin to the 1.2 ID now and test before the switch happens for you.

A storyboard test

Write a five-shot sequence with one recurring character. Supply three reference images. Generate each shot at 720p, then extend the best one twice. Score identity drift between shots, audio sync, and how many turns each shot needed. Multiply turns by the per-second price and you have a realistic cost per finished second, which is usually two or three times the headline figure.

Other generators are listed under text-to-video and image-to-video.

How to evaluate Gemini Omni 1.2 Flash

Compare the model on the work you need to complete.

Tasks and inputs

Gemini Omni 1.2 Flash is listed for Image-text-to-video, Image-to-video, Text-to-video, Video-to-video. It accepts text, image, video and produces video, audio, text. Use the publisher’s input format and preprocessing for your chosen deployment.

Capacity and cost

Documented limits: 131.07K context and 65.54K maximum output. The listed route has starting rates of $1.50 input and $12.00 output per million tokens. Choose duration, resolution and frame rate together. Compare the cost of an accepted clip, including repeat generations.

Choosing a deployment

Use the model identifier and account requirements for the selected API route. Confirm its request limits and output format with one complete task before connecting a larger workflow.

Frequently asked questions

Clips run 3 to 15 seconds. Each turn can extend a video by 3 to 10 seconds, up to 30 seconds in total.

These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.

pika

Pika H3 Flex

Pika H3 Flex is Pika's open LoRA post-training of MiniMax H3: six-step sampling, up to 15 s clips with 32 kHz stereo audio at 480p or 768p, 24 fps.

— context

heygen

HeyGen Video 1

HeyGen Video 1 generates 5 to 30 second MP4 clips with audio from text, images or reference media, up to 2K at 24 fps. Built on MiniMax H3.

— context

Meta

Muse Video

Muse Video is Meta’s previewed video-generation model with native audio; a public Meta Model API route and price have not been established.

— context

Abiray

Abiray/LTX-2.3-22B-DISTILLED-1.1-GGUF

Abiray/LTX-2.3-22B-DISTILLED-1.1-GGUF is a source-linked Hugging Face repository assigned to Text-to-video. Its recorded task interface accepts text and produces video.

— context