Updated

Google DeepMind

Veo 3.1

Google DeepMind's Veo 3.1: 8-second videos with dialogue, sound effects and ambience generated in one pass.

Create with Veo 3.1View pricing

What is Veo 3.1?

Veo 3.1 is Google DeepMind's video model, released on October 15, 2025.

It generates dialogue, sound effects and ambience in the same pass as the picture, so a scene with a spoken line and background traffic comes out already scored. On Flashloop it renders up to 8 seconds at 720p or 1080p in 16:9, with up to 3 reference images.

Audio is the reason most people pick it. With many video models you generate silent motion and then look for a voice or sound tool to finish the job; with Veo 3.1 you write the line in quotes, describe the room, and the clip arrives with both. It is also one of the better video models for short on-screen text, which matters for product labels, shop signs and app screens. On Flashloop it has passed 239,000 generations.

The limits are length and format. Clips stop at 8 seconds and, on Flashloop, render in 16:9 only, so vertical social video and longer takes are better served by Kling 3.0 or Seedance 2.5. Several people talking over each other can blur together, and hands in close-up are still hit and miss. For a single speaker, a product moment or a short cinematic beat with real sound, it is the first model to try.

Provider
Google DeepMind
Length
8 seconds
Resolution
720p, 1080p
Aspect ratios
16:9
Audio
Native audio

What you can do with Veo 3.1

  • Native audio

    dialogue, sound effects and ambience come out of the same render

  • Up to 3 reference images to guide a character, a product or a visual style

  • Readable text

    signs, labels and screens with a few words usually come out legible

  • 720p or 1080p output for social and ad exports

  • Widely used on Flashloop

    more than 239,000 videos generated

Prompts to try with Veo 3.1

“Close-up of a barista in a warm-lit cafe who smiles and says "Your order is ready," steam rising from the cup beside her. Soft espresso machine hiss and quiet chatter in the background. Warm 35mm film look, 16:9.”

Try this prompt

Where Veo 3.1 fits best

  • Ad concepts with a spoken line

    Write the line into the prompt and Veo 3.1 renders the voice, the lip movement and the room tone together, useful for testing a script before a real shoot.

  • Product demos with on-screen text

    Labels, packaging and UI screens stay readable, so an 8 second product clip can carry real copy.

  • Short cinematic beats with ambience

    Footsteps, wind, traffic and room tone render with the picture, which makes mood pieces feel lived-in without a sound pass.

  • Consistent subjects across a batch

    Up to three reference images keep a character's face, wardrobe or a product's shape steady across several generations.

How Veo 3.1 compares

Veo 3.1 compared with other video models on Flashloop
ModelLengthTop resolutionAspect ratiosNative audio
Veo 3.18 seconds1080p16:9Yes
Kling 3.015 secondsPro (1080p)16:9, 9:16, 1:1No
Kling 3.0 Turbo15 seconds1080p16:9, 9:16, 1:1No
Kling 4.0Coming soon30 secondsTo be confirmedTo be confirmedYes
Kling 2.610 seconds1080p16:9, 9:16, 1:1Yes

Good to know before you generate

  • Clips stop at 8 seconds, so chain clips for longer scenes
  • Several characters talking over each other can come out muddy
  • On Flashloop it renders in 16:9 only
  • Hands and fingers are still inconsistent in close-ups
  • Reference images and first or last frames cannot be combined in one generation

Start creating with Veo 3.1

Create with Veo 3.1

  1. Select Veo 3.1

    Open Flashloop and choose Veo 3.1 from the model selector in the video creator.

  2. Enter your prompt

    Describe what you want to see in your video.

  3. Generate and download

    Hit generate and your video will be ready in seconds. Download or share directly.

Questions about Veo 3.1

Does Veo 3.1 generate audio natively?

Yes. It generates dialogue, sound effects and ambience in the same pass as the video, so there is no separate voice or sound step.

What video length and resolution does Veo 3.1 support on Flashloop?

Clips up to 8 seconds at 720p or 1080p, in 16:9. For vertical video, pick a model that offers 9:16, such as Kling 3.0.

Can I use reference images with Veo 3.1?

Yes, up to 3, to guide the look of a character, an object, a product or the scene. Google calls this Ingredients to Video. References cannot be combined with first or last frames in the same generation.

Is Veo 3.1 good for scenes with text on screen?

It is one of the better video models for short text such as signs, labels and phone screens. Keep the text to a few words and put it in quotes.

What makes Veo 3.1 different from other AI video models?

Mainly the audio. Many video models still produce silent clips, while Veo 3.1 generates the picture, the dialogue and the environmental sound together.

What is Veo 3.1 best used for?

Short clips where sound matters: dialogue scenes, ad mockups, cinematic moments and product shots with readable text.

Who makes Veo 3.1?

Google DeepMind. Veo 3.1 follows Veo, Veo 2 and Veo 3, and Google also offers it in the Gemini app, Flow and the Gemini API. Flashloop gives you access to it alongside other video models.

Can Veo 3.1 generate vertical or square video on Flashloop?

Not on Flashloop, where it renders in 16:9 only. For 9:16 or 1:1, use Kling 3.0 or another model that lists those ratios.

More video models on Flashloop

Browse all models