The complete step-by-step guide to using Google's newest AI video model — from free to pro tiers.
By the ApiPass Editorial Team
Introduction
Google's Gemini Omni is the company's most ambitious multimodal AI model yet — a system that turns text, images, audio, and video into editable, physics-aware video, all through natural conversation. Whether you're a casual creator looking to spice up a YouTube Short or a professional filmmaker building a cinematic scene, there are three official ways to access Gemini Omni today, plus a fourth on the horizon for developers.
This guide walks you through each method step by step, including pricing and direct links to get started.
Method 1: Access Gemini Omni via the Gemini App
The Gemini app (available on web, iOS, and Android) is the primary and most full-featured way to use Gemini Omni. It supports the complete conversational editing experience — multi-turn refinements, multi-modal references (image + audio + video), and AI avatar generation.
Step-by-Step Instructions
- Open the Gemini app. Visit gemini.google.com in your browser, or download the Gemini app for iOS or Gemini app for Android.
- Sign in with your Google account that has an active Google AI Plus, Pro, or Ultra subscription.
- Enter video generation mode. In the prompt box, click the "+" icon and select "Create video" from the menu. This activates Gemini's video generation mode — Gemini Omni is already set as the default model under the hood.
- Confirm you're in video mode. You should now see a "Videos" tag appear inside the prompt input box, indicating that your next prompt will produce a video output.
- Type your prompt describing the video you want (e.g., "Create a 10-second clip of a violinist performing in a meadow at sunset").
- Attach reference inputs (optional). Use the attachment icon to upload reference images, audio clips, or existing video files — Omni can combine any mix of these modalities as references for a single generation.
- Send the prompt. Gemini Omni Flash will generate a video, typically within 30–90 seconds.
- Edit conversationally. Reply with follow-up instructions like "Make the background a snowy mountain" or "Rotate the camera to the back of the performer." Each instruction builds on the last while preserving previous edits.
- Download the result. Click the download icon on the generated video. All outputs include an imperceptible SynthID watermark.
Pricing
The Gemini app requires a paid Google AI subscription:
- Google AI Plus — Entry-tier subscription with limited Gemini Omni Flash access.
- Google AI Pro — $20/month, full access to Gemini Omni Flash + 200 Flow Credits.
- Google AI Ultra — $100/month (new tier) or $200/month (top tier), highest usage limits + 10,000–25,000 Flow Credits.
💡 Best for: Creators who want the full conversational editing experience with maximum flexibility.
Method 2: Access Gemini Omni via Google Flow
Google Flow is Google's dedicated AI creative studio, purpose-built for filmmakers and storytellers. It offers a more cinematic, scene-based workspace with timeline-style controls, making it the preferred environment for narrative video projects.
Step-by-Step Instructions
- Navigate to Google Flow and sign in with your Google AI–subscribed account.
- Create a new project by clicking "New Project" in the dashboard.
- Locate the model selector. Inside your project, find the model and generation-mode selector at the bottom-right corner of the prompt input box.
- Set generation mode to "Video." Click the generation-mode option and choose Video to switch Flow into video generation mode.
- Select Gemini Omni Flash. Open the model dropdown menu at the very bottom of the selector panel — you'll see that Omni Flash has been added to the available model list. Select it as your generation model. (Flow also still supports Veo 3.1 for longer-form output if you prefer.)
- Build your scene. Upload reference images, audio tracks, or existing video clips into the project library.
- Write your scene prompt in the prompt panel. Flow lets you organize multiple scenes side-by-side, making it easier to maintain visual consistency across a longer story.
- Generate the clip by clicking "Generate." Each generation consumes Flow Credits from your subscription quota (see pricing breakdown below).
- Iterate scene-by-scene. Use Flow's scene-stitching tools to chain multiple Omni-generated clips into a longer sequence.
- Export your project as a single video file (MP4) once you're happy with the cut.
Pricing
Google Flow is bundled with Google AI subscriptions:
- Google AI Pro ($20/month) — 200 Flow Credits per month.
- Google AI Ultra ($100/month or $200/month) — 10,000 or 25,000 Flow Credits per month, plus 5× higher usage limits.
Gemini Omni Flash credit consumption in Flow is based on the length of the generated clip:
| Clip Length | Flow Credits |
|---|---|
| 4 seconds | 15 credits |
| 6 seconds | 20 credits |
| 8 seconds | 25 credits |
| 10 seconds | 30 credits |
For example, a Google AI Pro subscriber with 200 monthly credits can generate roughly 6–7 ten-second clips per month, while an Ultra subscriber with 25,000 credits can produce hundreds of clips for larger projects.
💡 Best for: Filmmakers and creators building multi-scene cinematic projects.
Method 3: Access Gemini Omni via YouTube Shorts Remix (Free)
For casual creators who want to try Gemini Omni without a subscription, Google has integrated it into the Omni Remix feature on YouTube Shorts (and the YouTube Create app). Unlike the Gemini app, which generates videos from scratch, Omni Remix lets you transform existing Shorts by applying a new aesthetic via natural-language prompts — powered entirely by Gemini Omni.
Step-by-Step Instructions
- Find an eligible Short. Open YouTube (web or mobile app) and scroll through your feed to find a Short you want to remix. You'll know a video is eligible if the standard remix button is visible on the screen. Note that original creators can choose to opt out of visual remixing at any time.
- Tap "Remix." Tap the remix button located in the bottom right corner of the Short.
- Select "Re-imagine." From the drop-down menu, choose Re-imagine — this is the option that activates the generative AI powered by Gemini Omni.
- Enter your prompt. Type a custom text prompt describing the aesthetic you want — for example, "make this look like a high-budget sci-fi movie trailer" or "a grainy '90s horror film". You can also choose from one of YouTube's dynamic suggested prompts.
- Add reference photos (optional). To give the AI a precise visual target, upload reference photos straight from your camera roll. These act as image prompts that guide the color grading and visual style.
- Render the video. Tap Next and wait a few seconds. Gemini Omni will analyze the original frames and seamlessly rebuild the video and audio to match your prompt.
- Edit and publish. After the AI transformation, use YouTube's native Shorts editing tools to trim the timeline, add trending text, or adjust audio levels — then publish to your audience.
💡 Note: The same Omni Remix feature is available in the YouTube Create app, though the exact button placement may differ slightly from the YouTube Shorts interface.
Pricing
Free. Omni Remix is rolling out at no cost in both the YouTube Shorts Remix interface and the YouTube Create app. Google has also confirmed it's coming soon to the AI Playground feature for additional creative possibilities.
💡 Best for: Casual creators and social-media users who want to instantly restyle existing Shorts without paying for a subscription.
⚠️ Things to Keep in Mind When Using Omni Remix
- Creator controls and opt-outs: Not every Short is eligible for remixing. Original creators maintain full control over their content and can opt out of visual remixing at any time. You can only use Omni Remix if the standard remix button is visible on the video.
- Automatic attribution: When you publish a remixed video, YouTube automatically adds a clickable attribution link back to the original creator's video, ensuring they receive fair credit and traffic.
- AI transparency: All Shorts remixed via Gemini Omni automatically include digital watermarks (SynthID) and identifying metadata to clearly indicate AI involvement.
- Likeness protection: YouTube is expanding its industry-first Likeness Detection tool to all creators aged 18 and older, helping them detect and manage how their likeness is used across the platform.
- Cost and future availability: Omni Remix is currently free in both YouTube Shorts and the YouTube Create app, with broader rollout to the AI Playground coming soon.
Extra: An Upcoming Way — Using Gemini Omni via Developer API
Currently, Gemini Omni does not have a public API. However, per Google's official launch announcement, Google has confirmed that API access for developers and enterprise customers will roll out in the coming weeks.
When it launches, the Gemini Omni API will likely be available through:
- Google AI Studio — for rapid prototyping and testing.
- Vertex AI — for production-grade enterprise deployments on Google Cloud.
- Gemini API — for direct integration into apps via REST/SDK calls.
Pricing for the Gemini Omni API has not yet been announced. Based on Veo 3.1 API pricing as a reference, expect per-second video generation pricing in the range of $0.10–$0.50 per second depending on resolution and tier.
Get Gemini Omni Through ApiPass (Coming Soon)
At ApiPass, we're already preparing to bring Gemini Omni to our developer community the moment Google opens public API access. Our goal is to give you one unified endpoint to access the entire Gemini ecosystem — alongside hundreds of other models — without juggling separate provider accounts.
In the meantime, if you need production-grade AI video generation today, ApiPass already offers the full Veo 3.1 family in three tiers:
- Veo 3.1 Lite — The most cost-efficient tier, ideal for high-volume generation and prototyping.
- Veo 3.1 Fast — Optimized for throughput and rapid iteration, balancing speed and quality.
- Veo 3.1 Quality — The premium tier for the highest visual fidelity and cinematic output.
The moment Google opens the Gemini Omni API, ApiPass will integrate it shortly after its official release, giving you immediate developer access through the same unified endpoint you already use for Veo 3.1.
💡 Best for: Developers and businesses building AI video features into their own apps, products, or automated pipelines.
Quick Comparison: Which Method Should You Choose?
| Method | Price | Best For | Output |
|---|---|---|---|
| Gemini App | $20–$200/month | Full conversational video creation & editing | 10s generated clips |
| Google Flow | $20–$200/month | Cinematic multi-scene projects | 4–10s clips (15–30 credits each) |
| YouTube Shorts / Create App (Omni Remix) | Free | Restyling existing Shorts via prompt | Remixed Short |
| Developer API (upcoming) | TBD | Building AI video into your own apps | TBD |
Conclusion
Gemini Omni is one of the most accessible AI video models ever launched — Google has made it free for casual creators through YouTube's Omni Remix feature while reserving the deepest creative capabilities for paid Gemini app and Google Flow subscribers. Whichever method you choose, you're tapping into the same core engine: a physics-aware, conversationally controllable video model that finally feels like collaborating with a creative partner.
For developers and businesses, the wait is almost over. Bookmark the ApiPass Veo 3.1 endpoints for production needs today — and stay tuned to our blog for the Gemini Omni API rollout, coming soon.
Welcome to the Omni era. 🚀