Veo 3.1
Cinematic scenes with character dialogue and synced sound
- Up to 3 reference images to keep the same character
- Smooth transition between a start and an end image
- With sound on and no reference images, 1080p and 4K are upscaled from a 720p clip
What you can make with Veo 3.1
- Ways to create
- Text to video
- Image to video
- References
- Length
- 4, 6, or 8 seconds
- Resolution
- 720p
- 1080p
- 4K
- Aspect ratio
- 16:9
- 9:16
- Sound
- Built in, can be turned on or off. On by default
- Files you can add
- Start image
- End image
- Reference images (up to 3)
- Text length
- Prompt: up to 10,000 characters
- More options
- Negative prompt
Price by settings
How to read the tableEach row and each column changes one setting, and everything else stays at the model’s default settings. The marked cell is the default settings, the price at the top of the page. "Not offered" is a combination the model does not offer right nowEach cell is one generation at its settings, in credits and in shekels with VAT included. Prices are computed when the page loads, with the same calculation the studio shows before a generation
| Duration (seconds) · Resolution | 720p | 1080p | 4K |
|---|---|---|---|
| 4s | 58.58 cr₪5.86 | 59.76 cr₪5.98 | 89.05 cr₪8.91 |
| 6s | 58.58 cr₪5.86 | 59.76 cr₪5.98 | 89.05 cr₪8.91 |
| 8s | 58.58 cr₪5.86Default settings | 59.76 cr₪5.98 | 89.05 cr₪8.91 |
Options that change the price
At the default settings, changing only this option
- Generate audio: Off74.99 cr₪7.50
Hebrew
No official confirmation of Hebrew speechWhen a model is marked Good HebrewWhen the model’s official language list includes Hebrew, when our own tests passed, or, for some image models, from our own experience using them. For every other model this says what is known about its Hebrew
Questions about Veo 3.1
What is Veo 3.1 and what is it best at?
Veo 3.1 is Google DeepMind’s video model for cinematic clips of 4 to 8 seconds, with dialogue, effects and ambience generated together with the picture. It is especially good for scenes with dialogue, smooth transitions between a start and an end image, and keeping a character or product the same with up to 3 reference images. You can make landscape 16:9 or vertical 9:16 clips at 720p, 1080p or 4K
How much does an 8 second Veo 3.1 clip cost?
Right now an 8 second clip with sound costs ₪5.8658.58 crVeo 3.1, 720p · 8s · with sound at 720p, ₪5.9859.76 crVeo 3.1, 1080p · 8s · with sound at 1080p and ₪8.9189.05 crVeo 3.1, 4K · 8s · with sound at 4K. The exact price shows before every generation, and the table on this page shows the price for every length and resolution. For cheaper drafts there are Veo 3.1 Fast and Veo 3.1 Lite
Does Veo 3.1 speak Hebrew?
There is no official confirmation. Google confirms full support for English and says other languages have not been evaluated and may work with varying results, and we have not tested Hebrew speech either. For a Hebrew voiceover it is safer to add a voice from a speech model that supports Hebrew, such as Eleven v3, and join them in an editor
How do I keep the same character in Veo 3.1?
Attach up to 3 reference images of the character, object or scene, and Veo 3.1 keeps them the same in the new clip. According to Google, reference images work in 8 second clips, and in this mode you do not add a start or end image. Use the same images and the same character description in every clip of a series
Is Veo 3.1 being replaced by Gemini Omni Flash?
Google announced that the Veo 3.1 preview versions in the Gemini API close on October 22, 2026, and pointed to Gemini Omni Flash 1.1 as the replacement. Google lists the stable Veo 3.1 for retirement no earlier than November 17, 2026. Both models are available here today, and files you already made stay in your library even if a model stops working
Guides that mention Veo 3.1
- What creating with AI costs and how to pay lessWhat an AI video, image or voiceover costs: what sets the price, live prices for common settings, subscription vs pay per generation and how to pay less
- How to make a video with AI: the complete guideHow to make a video with AI from scratch: choosing a model, text, image or sample images, every setting, a prompt that works, the cost and what to avoid
- Hebrew in AI models in 2026: what really worksWhat works in Hebrew in AI models today: voiceover, text in images, speech in video, prompts and songs, by official lists, our tests and what is unverified
- How to write an AI video prompt: structure and camera movesHow to write a prompt for an AI video: the official five part structure, dialogue and sound, a camera move glossary and which language to write in
- How to choose an AI video model for the jobHow to choose an AI video model: talking characters, a product that stays the same, length, resolution, price and Hebrew in Veo, Kling, Seedance and Wan
- How to keep the same character across videos and imagesHow to keep the same character or product across videos and images: reference images, start images, how many each model takes, a workflow and live prices
- AI video longer than 8 seconds: models that reach 30How long an AI video can be: which models make more than 8 seconds in one generation, up to 15, 20 and 30 seconds, what it costs and how to go longer
- Sora shut down: what happened and what to use insteadOpenAI discontinued Sora: the app closed in April 2026 and the API in September 2026. What happened, what happens to your videos and what to use now
- Veo 3.1 vs Kling 3.0: which is better and whenVeo 3.1 vs Kling 3.0: clip length, sound and dialogue, resolution, start images and references, independent rankings and live shekel prices
- Veo 3.1 vs Veo 3.1 Fast vs Veo 3.1 Lite: which to chooseHow the three versions of Google’s Veo 3.1 differ: length, resolution, sound, reference images, independent rankings and live shekel prices
Similar models
- Veo 3.1 FastShort clips with dialogue and sound, from the Veo 3.1 familyVideo from text, an image, or reference images, up to 4K14.06 cr₪1.41 for a 8s clip at 720p
- Veo 3.1 LiteMany short clips with sound at a low costVideo from text, an image, or reference images, up to 4K7.03 cr₪0.70 for a 8s clip at 720p
- Wan 3.0 PrimeThe same abilities as Wan 3.0 with a shorter wait, at a higher priceVideo from text, an image, or reference images, up to 1080p59.05 cr₪5.91 for a 5s clip at 1080p
- Seedance 2.5Clips up to 30 seconds in one generation with soundVideo from text, an image, or reference images, up to 1080p54.18 cr₪5.42 for a 5s clip at 720p
- LTX-2.5 ProThe quality mode of LTX-2.5 for 6, 8 or 10 second clips with soundVideo from text or an image, up to 1080p47.81 cr₪4.78 for a 6s clip at 1080p
- Seedance 2.0Scenes with complex motion like sports and action, with soundVideo from text, an image, or reference images, up to 4K42.18 cr₪4.22 for a 5s clip at 720p