Creating a short AI video often means balancing clip length, editing control, and cost. With its latest update, Gemini Omni 1.2 Flash adds support for clips of up to 15 seconds, audio generation, and prompt – based editing with reference materials.
In this guide, we’ll look at what’s changed from version 1.1, how the features work in practice, what the model costs, and how to try it on ChatGOAT AI.
What Is Gemini Omni 1.2 Flash?
Gemini Omni 1.2 Flash is Google’s AI model for creating and editing short videos. It can take text, image, and video inputs to generate new clips, edit or extend existing ones, and improve video quality. Unlike text – only Gemini models, it can also generate audio to accompany the video.
This gives creators a way to work on visuals and sound in the same workflow. Google says each generated video also includes an invisible SynthID watermark to identify it as AI – generated.
What’s New in Gemini Omni 1.2 Flash?
15-Second Video Generation with Audio
One of the main updates in Gemini Omni 1.2 Flash is support for video clips ranging from 3 to 15 seconds, adjustable in one – second increments. Gemini Omni 1.1 was limited to 10 seconds, so the longer maximum gives creators more room to show an action or develop a short scene.
The model can also generate audio alongside the video, including speech, music, and sound effects. This can make clips more complete for social posts, demos, or early – stage concepts. Using uploaded audio as a reference requires Google Cloud permissions and isn’t currently available through the public Gemini API.
Video Editing with Reference Images and Text Prompts
Gemini Omni 1.2 Flash lets users guide video edits with reference images, short video clips, and text prompts. These inputs can help shape a clip’s visual style, composition, or motion.
After generating a video, users can make changes with follow – up instructions – for example, changing a character’s outfit, warming the lighting, replacing an object, or applying a watercolor look. This makes it easier to refine a 15 – second clip without starting over for every adjustment.
Camera Movement and Scene Control
Users can specify camera movement in their prompts to guide how a shot unfolds – for example, requesting a slow pan, gradual zoom, or tracking shot. They can also describe the desired framing, such as a wide shot that shows more of the surroundings. Clear directions help communicate the intended look and movement, though the final result may vary.
Low – Resolution Drafts for Faster Iteration
Gemini Omni 1.2 Flash supports video generation at 360p and 720p, with options to upscale the output later. Google’s documentation lists these upscaling paths:
- 360p to 720p or 1080p
- 720p to 1080p or 4K
Starting with a lower – resolution draft can help you review a clip’s content and make changes before upscaling it. This can be useful when testing ideas, though the time or cost savings will depend on the workflow and pricing.
Gemini Omni 1.2 Flash VS. Gemini Omni 1.1: What’s Different?
| Feature | Gemini Omni 1.1 Flash | Gemini Omni 1.2 Flash |
| Maximum clip length | Up to 10 seconds | Up to 15 seconds |
| Duration options | 3 – 10 seconds | 3 – 15 seconds, with 10 seconds as the default |
| Video extension | Up to 10 seconds per turn; up to 40 seconds total | 3 – 10 seconds per turn; up to 30 seconds total |
| Editing reach | Edits to the same generated take | Edits can apply across the whole clip, including its aspect ratio |
| Reference inputs | Up to 10 images; up to 3 reference videos | Up to 6 images; reference – video options depend on output length |
| 720p API pricing | About $0.10 per second | About $0.07 per second |
| Watermark | SynthID | SynthID |
| Availability on Morphic | Available in the video model picker | Not yet available |
Output Length and Generation Quality
Gemini Omni 1.2 Flash supports clips of up to 15 seconds, giving users more room to show an action or develop a short scene. Longer output does not guarantee better quality, but it can reduce the need to combine multiple clips.
Editing Precision and Consistency
Version 1.2 Flash supports reference images, short videos, and follow – up prompts for editing. These options make it easier to refine a clip, though results can vary depending on the prompt and source material.
Generation Speed and Cost Differences
Gemini Omni 1.2 Flash costs $0.07 per second for 720p generation – about 30% less than Gemini Omni 1.1. The lower price can make it more affordable to test prompts or generate multiple versions.
What Can You Do with Gemini Omni 1.2 Flash?
Preserving Motion and Sound from a Source Clip
A source clip can guide the movement in a generated video, while a prompt describes changes to the scene and the audio. For example, you could use footage of someone walking as a motion reference, ask for a rainy street setting, and request rain and footsteps as generated sounds. Review the result to check that the movement and audio fit the scene.
Replacing One Object While Keeping the Background
Changing a single object in a video can be tricky if the rest of the scene needs to stay intact. With prompt – based editing, you can ask Gemini Omni 1.2 Flash to replace a specific item while keeping the surrounding setting – for example, swap a blue mug on a table for a glass while retaining the table, room, and camera angle. Check the edited clip to make sure the replacement remains consistent as the scene moves.
Applying a Full Style Change Across a Moving Scene
Gemini Omni 1.2 Flash allows users to change the visual style of an existing video through follow – up prompts while keeping the main action and scene composition. For example, you can ask Gemini Omni 1.2 Flash to:
- Turn a realistic scene into a stop – motion animation
- Transform a video into a watercolor or cartoon style
- Apply a cinematic look with different lighting and color tones
These edits can give an existing clip a completely different visual appearance without requiring you to generate an entirely new scene. However, some details may not remain consistent across frames, so review the edited video to make sure the new style is applied as intended.
Generation Time and Cost Observations
Generating a video at a lower resolution first can help refine your ideas before producing the final version. Gemini Omni 1.2 Flash supports 360p and 720p generation, with upscaling options up to 1080p or 4K.
For example, you can create a 360p draft to check the composition and camera movement, then adjust your prompt before generating or upscaling the final video. This approach can help avoid unnecessary high – resolution processing, though actual time and cost savings depend on usage and pricing.
Gemini Omni 1.2 Flash Pricing: What It Costs to Use
Gemini Omni 1.2 Flash’s listed rate for 720p video generation is $0.07 per second. The total cost may vary depending on the clip length, resolution, and any additional processing or upscaling.
If you’d like to check the latest pricing and billing details, visit Google’s official pricing page. Rates may change over time, so it’s worth checking before starting a project.
How to Use Gemini Omni 1.2 Flash on ChatGOAT AI
ChatGOAT AI not only brings a range of advanced AI models together in one place, making it easier to try different prompt styles, but also lets you try Gemini Omni 1.2 Flash for free – without setting up a complex workflow yourself.
For users interested in AI video creation, ChatGOAT AI makes it easier to test the model and compare results. You can see whether its longer video clips and editing controls suit your needs, then refine your prompts based on what works. For creators who want to test, compare, and iterate on one platform, it’s a straightforward way to begin.
Final Verdict: Is Gemini Omni 1.2 Flash Worth Using?
Gemini Omni 1.2 Flash offers a more flexible workflow than Gemini Omni 1.1, particularly for short videos that need a few rounds of editing. Its longer clip limit, audio generation, reference inputs, and lower listed 720p price make it worth considering for drafts and simple variations.
It may be a good fit if you want to:
- Generate clips of up to 15 seconds rather than 10
- Add generated speech, music, or sound effects
- Make targeted edits with reference materials and follow – up prompts
- Test ideas at lower resolutions before upscaling
Keep in mind that follow – up edits may affect timing, motion, or continuity. Review each version carefully, as repeated generations can increase costs.
Overall, Gemini Omni 1.2 Flash is a good fit for short – form projects that benefit from quick drafts and revisions. Try it on ChatGOAT AI to see if it suits your workflow.