How to Use Google Gemini Omni Flash 1.1 on VdoBloom
Gemini Omni Flash 1.1 is the one Gemini Omni model that runs both text-to-video and image-to-video: real specs, the full 360p to 4K credit table, prompting notes and when to pick another tier.
Google Gemini Omni Flash 1.1 is the fast tier of the Gemini Omni family on VdoBloom, and its defining trait is that one model id does both text-to-video and image-to-video — 4, 6, 8 or 10 second clips at 360p, 720p, 1080p or 4K in 16:9 or 9:16, starting at 36 credits ($1.80) for a 4-second render and topping out at 120 credits ($6.00) for 10 seconds at 4K. It is the only member of the family that adds a 360p tier, and the only one you never have to swap out when you move between a written prompt and a reference photo.
What makes Flash 1.1 different from the rest of the family
The Gemini Omni family on VdoBloom has three generation members: a text-to-video model, an image-to-video model, and Flash 1.1. The first two are locked to one workspace tab each. Flash 1.1 is valid in both the Text-to-Video and Image-to-Video tabs, which changes how you work with it.
In practice a single settings block carries across a whole session. You write a prompt, render it, like the framing, then upload a still and animate that instead — without re-picking a model, re-checking which durations it allows, or re-learning a different resolution list. For anyone building a batch of clips where some start from text and some start from product photos, this is the model that keeps the workflow in one place.
The trade-off is what Flash 1.1 will not do: it rejects video input outright. If you want to transform an existing clip rather than generate a new one, Flash is the wrong door.
Real specs, side by side with its sibling
| Spec | Gemini Omni Flash 1.1 | Gemini Omni Text-to-Video |
|---|---|---|
| Capability | Text-to-video AND image-to-video | Text-to-video only |
| Durations | 4, 6, 8, 10 seconds | 4, 6, 8, 10 seconds |
| Resolutions | 360p, 720p, 1080p, 4K | 720p, 1080p, 4K |
| Aspect ratios | 16:9, 9:16 | 16:9, 9:16 |
| Reference images | Up to 7 | Up to 7 |
| Character references | Up to 3 | Up to 3 |
| Video input | Not supported | Not supported |
| Content filter | STRICT | STRICT |
| Family rating | 4.7 | 4.7 |
The combined input quota is worth reading carefully: images plus character references must total 7 or fewer, and no more than 3 of those may be character references. Six product photos plus one registered character is fine; five characters is not.
What every tier costs, in credits and dollars
Flash 1.1 shares the family credit table, with the 360p column added. One credit is $0.05 at the Lite rate ($15 for 300 credits), so these dollar figures are the real cost of a single render on that plan.
| Duration | 360p | 720p | 1080p | 4K |
|---|---|---|---|---|
| 4 seconds | 36 cr — $1.80 | 36 cr — $1.80 | 36 cr — $1.80 | 84 cr — $4.20 |
| 6 seconds | 48 cr — $2.40 | 48 cr — $2.40 | 48 cr — $2.40 | 96 cr — $4.80 |
| 8 seconds | 60 cr — $3.00 | 60 cr — $3.00 | 60 cr — $3.00 | 108 cr — $5.40 |
| 10 seconds | 72 cr — $3.60 | 72 cr — $3.60 | 72 cr — $3.60 | 120 cr — $6.00 |
Read that table honestly, because it contradicts the obvious assumption: 360p, 720p and 1080p all bill at exactly the same credit price. The 360p tier is a speed and file-size option, not a discount. There is no scenario in which choosing 360p saves you a credit, so pick 1080p for anything you might keep and reserve 360p for throwaway motion tests where you only want to see whether the action reads.
Only 4K carries a real premium: an extra 48 credits ($2.40) at every duration. Because that surcharge is flat while the base price climbs with length, 4K is proportionally cheapest at 10 seconds — 1.7× the 1080p price — and most expensive at 4 seconds, where it is 2.3×.
New accounts get 10 free credits with no card, which will not cover a Flash render, so plan on either a credit pack from $2.49 for 75 credits that never expire, or a subscription from $15 per month. Paid plans also download watermark-free with commercial rights. The full breakdown lives on the pricing page.
How to use it: text first, then image
- Open the Text-to-Video workspace and select Gemini Omni, then the Omni Flash 1.1 variant.
- Set duration and resolution. For a first pass, 4 seconds at 1080p costs 36 credits and tells you whether the concept holds.
- Write the prompt as a shot: subject, one action, camera move, light, location. Add spoken lines in quotes if you want native dialogue.
- Render, then judge only two things — does the action read, and is the framing right. Ignore small artefacts on a draft.
- To animate a photo instead, switch to the Image-to-Video workspace with Flash 1.1 still selected and upload your still. Durations, resolutions and prices are unchanged.
- Re-render the keeper at 1080p or 4K at your final duration.
If your reference photo shows a real person, you must have that person’s consent before animating them. That is a platform rule, not a suggestion, and the STRICT content filter on this family independently rejects sensitive likeness prompts.
Prompting notes specific to Flash
Fast tiers generally trade some prompt adherence for speed, and Flash 1.1 behaves accordingly: it is more literal and less interpretive than the full text-to-video model. Three adjustments help.
- Shorten the prompt. Long, layered descriptions get partially dropped. Two or three tight clauses land better than a paragraph.
- Name the camera move explicitly. Slow push-in, locked-off wide, handheld follow. Without one, Flash tends to default to a drifting camera.
- State what stays still. Like the rest of the family, Flash will animate everything in frame unless a clause pins the background down.
On image-to-video, keep the prompt to the motion only. The photo already settles wardrobe, lighting and composition; re-describing them invites the model to redraw what you liked.
When to choose something else
- Maximum prompt fidelity from text. Go to Gemini Omni Text-to-Video, the full-weight sibling at identical prices.
- A dedicated photo-animation run. Gemini Omni Image-to-Video is the specialised tier for stills.
- Clips longer than 10 seconds. Kling 3.0 Omni Text-to-Video runs 3 to 15 seconds in a single shot.
- Flexible content filtering. Flash is STRICT. Swimwear, dance and fitness concepts belong on the flexible-tier models described on the flexible-filter generator page; explicit and illegal content is blocked everywhere.
Frequently asked questions
Is Gemini Omni Flash 1.1 cheaper than the other Gemini Omni models?
No. It shares the family credit table exactly: 36, 48, 60 and 72 credits for 4, 6, 8 and 10 seconds up to 1080p, and 84 to 120 credits at 4K. What it adds is speed and a 360p option, not a lower price.
Then what is the 360p tier for?
Quick motion checks and small file sizes. Since it costs the same as 1080p, use it only when you genuinely want a fast, low-weight draft rather than a keeper.
Does it do image-to-video?
Yes. Flash 1.1 is the one Gemini Omni id valid in both the Text-to-Video and Image-to-Video tabs, and it accepts up to seven reference images.
Can I feed it an existing video?
No. Video input is rejected for Flash 1.1. Video-to-video transformation is handled by a different model in the family inside the edit workspace.
Am I charged if the render fails?
No. Credits are deducted when the job is submitted and refunded automatically on failure, with a safeguard that blocks a refund if the video actually completed.
What aspect ratios are available?
Only 16:9 and 9:16. If you need a square 1:1 output, generate 16:9 and crop, or choose a model that supports 1:1 natively.
Ready to try it?
Create your first AI video in minutes — no credit card required.
Start Creating Free →