Gemini Omni Flash is Google’s current video model, the one Google names as Veo 3.1’s replacement: text, image or references to video, from 360p drafts to 4K.
Gemini Omni Flash is Google’s multimodal video model and the successor Google points Veo 3.1 users to. It generates from a text prompt, animates a still with an optional end frame, and holds a subject from up to four reference images, in shots of 3 to 10 seconds.
The resolution ladder is the useful part: 360p renders cost very little and are good enough to test a prompt, then the same prompt re-renders at 720p, 1080p or 4K for the final. A companion edit model, Gemini Omni Edit, re-renders footage you already have from a description.
From 19 tokens ($0.19) per clip; 141 tokens ($1.41) at the default settings.
| Settings | Tokens | Price |
|---|---|---|
| 3 s · 360p | 19 | $0.19 |
| 3 s · 720p | 56 | $0.56 |
| 3 s · 1080p | 82 | $0.82 |
| 3 s · 4k | 157 | $1.57 |
| 8 s · 360p | 47 | $0.47 |
| 8 s · 720p | 141 | $1.41 |
| 8 s · 1080p | 207 | $2.07 |
| 8 s · 4k | 401 | $4.01 |
| 10 s · 360p | 58 | $0.58 |
| 10 s · 720p | 174 | $1.74 |
| 10 s · 1080p | 256 | $2.56 |
| 10 s · 4k | 498 | $4.98 |
Prices come from the live catalog at $0.0100 per token, the pay-as-you-go rate; subscriptions include tokens at a lower rate. The exact price is shown before you generate, and a failed generation is refunded.