# Veo 3.1 prompt guide

URL: https://dreambench.ai/guides/veo-3-1-prompt-guide

To write a Veo 3.1 prompt, describe one shot in Google’s order: the camera and framing, the subject, the action, the setting, then the style, light and mood. Then say what is heard: dialogue in quotes after its speaker, effects after “SFX:”, the background after “Ambient noise:”. Keep the action to 4, 6 or 8 seconds.

Written and reviewed by the DreamBench team. Published 2026-10-31; updated 2026-10-31.

## Veo 3.1 prompts at a glance

- **The formula:** Cinematography, subject, action, context, style and ambiance, then the sound.
- **One generation:** 4, 6 or 8 seconds, 16:9 or 9:16, always with sound.
- **Sound cues:** Dialogue in quotes after its speaker, “SFX:” for effects, “Ambient noise:” for the background.
- **Prompt length:** 50 to 150 words for most clips; up to 2,600 characters on DreamBench.
- **On DreamBench:** Veo 3.1, Fast and Lite on the Video page, your prompt sent as written; 4K on Veo 3.1.

The [Veo 3.1 prompt generator](https://dreambench.ai/tools/veo-prompt-generator) writes a prompt in this order; this guide explains each part. For the model, see the [Veo 3.1 page](https://dreambench.ai/models/veo-3.1).

## Google’s Veo 3.1 prompt formula, part by part

Google’s [prompting guide for Veo 3.1](https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-veo-3-1) gives five parts. The camera comes first because it decides what the rest must fill; the sound follows in sentences of its own.

**Cinematography → subject → action → context → style and ambiance → sound**

|  | Why it matters | Strong | Weak |
|---|---|---|---|
| Cinematography | The shot sets what is in frame for every second. | Low-angle tracking shot, 35mm lens. | Cinematic shot. |
| Subject | The details you name are the ones Veo holds. | A cyclist in a mustard rain jacket and a white helmet | A cyclist |
| Action | One verb with an end gives the clip a shape. | brakes at the curb, unclips one foot and looks back | riding |
| Context | Place, time and weather stop Veo inventing them. | on a wet cobbled bridge in Amsterdam at dusk | in a city |
| Style and ambiance | A light source can be rendered; praise can’t. | Photoreal, cool dusk, warm shop light from the left. | Beautiful, 8K, masterpiece. |
| Sound | Veo always makes sound; unnamed, it chooses. | SFX: brakes squeal. Ambient noise: tires on wet stone. | Nice audio. |

1. Open with the camera: shot size, angle, one move and, if it matters, the lens.
2. Name the subject with the details that must not change.
3. Write one action in the present tense, from start to end.
4. Add where and when, the weather and the background.
5. Set the look, the light and its direction, and the mood.
6. End with the sound: any line in quotes after its speaker, then “SFX:” and “Ambient noise:”.

Google DeepMind’s guide for [Veo 3](https://deepmind.google/models/veo/prompt-guide/) uses the same parts and shows each prompt beside its video, from one line to a long play-by-play. A prompt written for the older model works on 3.1.

## Veo 3.1 camera movement, angle and lens prompts

Veo reads film language, so name the shot as a camera operator would. These terms come from Google’s [prompt guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/video-gen-prompt-guide).

**Camera words for Veo 3.1, and how to write them**

|  | What it does | How to write it |
|---|---|---|
| Static shot | Doesn’t move; the action carries it. | Static shot, medium close-up. |
| Dolly | Travels toward or away. | Slow dolly in to a close-up. |
| Tracking shot | Travels with the subject. | Tracking shot alongside her. |
| Pan and tilt | Turns from one spot. | Slow pan left across the harbor. |
| Crane shot | Rises or falls to reveal. | Crane shot rising to reveal the canyon. |
| Aerial view | Seen from the air. | Aerial drone shot over the islands. |
| Arc shot | Circles the subject. | Arc shot, a half turn round her. |
| Handheld or POV | A human shake, or a character’s eyes. | POV shot of hands on the wheel. |
| Low or high angle | Makes a subject loom or look small. | Low-angle shot of the statue. |
| Wide-angle or telephoto | Takes in the place, or isolates a subject far off. | Telephoto lens from across the street. |
| Macro lens | Fills the frame with something tiny. | Macro lens on a bee on lavender. |
| Shallow depth of field | The subject sharp, the background soft. | 85mm lens, shallow depth of field. |
| Rack focus | Focus shifts mid-shot. | Rack focus from the ring to her face. |

One move per clip: a dolly that becomes an arc in 8 seconds smears both. For dialogue, a static shot or a slow dolly in keeps the face steady.

## How to make Veo 3.1 talk: dialogue, sound effects and ambience

Every Veo 3.1 clip has sound, made with the picture. Google asks you to [describe it in sentences of its own](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/video-gen-prompt-guide), and its [Veo 3.1 guide](https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-veo-3-1) labels each layer. See [native audio](https://dreambench.ai/glossary/native-audio).

1. Name the speaker, then the line in quotes: A woman says, “We have to leave now.”
2. Give a tone where it matters: he says in a weary voice, “Not again.”
3. Put each effect after “SFX:”: SFX: the door slams.
4. Put the background after “Ambient noise:”: Ambient noise: the hum of a starship bridge.
5. Write music as a sound: a gentle orchestral score begins to play.

- One short line in 4 seconds, two in 8: speech runs at two to three words a second.
- Give every line a speaker in shot, or it can become a voice-over.
- Write lines in English: Google’s [model reference](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-1-generate) lists English as the prompt language. Veo moves the lips in the same generation; see [lip sync](https://dreambench.ai/glossary/lip-sync).

#### Subtitles and garbled speech

Google lists [natural speech in short lines](https://deepmind.google/models/veo/) as a limit it is still working on. [MIT Technology Review](https://www.technologyreview.com/2025/07/15/1120156/googles-generative-video-model-veo-3-has-a-subtitles-problem/) found garbled subtitles on clips with dialogue even when the prompt asked for none; Google’s advice was to run it again. So describe a clean frame, leave signs and captions out unless you want words on them, keep lines plain, and listen to every take.

## How long a Veo 3.1 prompt should be: what fits in 4, 6 and 8 seconds

The length is set before Veo reads the prompt, so write only what the seconds hold: a longer action is rushed, a longer line cut off.

**How much a Veo 3.1 clip holds**

|  | What fits | Prompt length |
|---|---|---|
| 4 seconds | One action and one sound, or a line of about eight words. | 30 to 60 words |
| 6 seconds | An action with a payoff, or a line and a reaction. | 50 to 100 words |
| 8 seconds | A build-up and payoff, two short lines, or four timed shots. | 60 to 150 words |

Google’s guide also splits one 8-second clip into timed shots, each opening with its range, such as [00:00-00:02]. Two seconds is short, so keep it for when the cuts matter. Veo makes 16:9 or 9:16 only; for square or 21:9, see the [Kling 3.0 prompt guide](https://dreambench.ai/guides/kling-3-0-prompt-guide).

## Veo 3.1 image-to-video and reference-image prompts

Google’s tools take pictures two ways: a first frame to animate, or up to three [reference images of one person, character or product](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/use-reference-images-to-guide-video-generation) to keep in a new shot. See [image to video](https://dreambench.ai/glossary/image-to-video) and [reference to video](https://dreambench.ai/glossary/reference-to-video).

**What to write when Veo 3.1 is given a picture**

|  | What the picture fixes | What the prompt says |
|---|---|---|
| A first frame | The subject, place, light and look. | Only what moves, the camera and the sound. |
| Reference images | What the subject looks like, not where it is. | The whole shot, naming the subject in words. |

Google’s reference example names its subject plainly: “A person walks in carrying a vase full of flowers and places the vase on a kitchen table.” References work on Veo 3.1 and Fast, not Lite, and the full model then makes 8 seconds.

On DreamBench your pictures become sheets. Veo 3.1 takes up to 3 with each clip, which then runs 8 seconds; on Fast and Lite, DreamBench draws an opening frame from them.

## First and last frame prompts in Veo 3.1

Google offers first and last frame on all three tiers: a start and an end picture, and Veo makes the move between them, with sound, at 4, 6 or 8 seconds ([Google Cloud](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-first-and-last-frames)). Describe the change, as Google’s example does: “a hand reaches in and places a glass of milk next to the plate of cookies”. Keep the frames close in light and framing. See [first and last frame](https://dreambench.ai/glossary/first-and-last-frame).

DreamBench doesn’t offer a start or end frame for Veo: its composer draws each clip from your references, so the prompt’s last sentence says how the shot ends.

## Veo 3.1 negative prompts: what works

Google’s API has a [negative prompt](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-text) field for what you don’t want. Its [prompt guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/video-gen-prompt-guide) says to list the things, “wall, frame”, never “no walls”; in the main prompt, its [Veo 3.1 guide](https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-veo-3-1) says to describe the result instead.

**Saying what you don’t want, in Veo’s terms**

|  | Instead of | Write |
|---|---|---|
| In the prompt | No man-made structures. | A desolate landscape with no buildings or roads. |
| In the prompt | No people. | An empty street at dawn. |
| In a negative prompt field | Don’t add text. | text, watermark |

DreamBench sends Veo no negative prompt field. The prompts it writes carry a short “Negative:” line naming burned-in subtitles where a clip speaks. On the Video page, describe the result and keep any exclusion to one closing line, as the [prompt generator](https://dreambench.ai/tools/veo-prompt-generator) does.

## Keeping a character consistent across Veo 3.1 shots

Two Veo clips agree only on what both were given. Google’s Veo 3.1 guide makes reference images of the cast and setting first. In the prompt:

- Describe each character in the same words every time: “a tall woman in her sixties with cropped silver hair and a green waxed jacket”.
- Keep the clothes, props, time of day, light and voice.

On DreamBench each character, place and prop is drawn once as a sheet, and every shot it is in starts from it. See [character consistency](https://dreambench.ai/glossary/character-consistency).

## Multi-scene Veo 3.1 videos: from one idea to a sequence

Google shows two ways past one shot: timed shots inside one 8-second clip, or separate clips made from the same references. DreamBench takes the second: a film is several Veo generations, one shot each, cut together, so it runs past 8 seconds and any shot can be redone.

*Diagram: One idea, several shots: each its own generation, hard cuts at the joins, any one redone alone.*

*Video (veo-3.1, 16:9, 1080p, 15 s, with sound): One brief, several clips on [Veo 3.1](https://dreambench.ai/models/veo-3.1): 16:9 at 1080p, 15 seconds with sound. The light and the muffled underwater sound carry across the cut.* [Watch](https://media.dreambench.ai/templates/tpl-g-p5iq7agjo4.mp4)

> Brief: Underwater at sunrise: light shafts sweep across a coral reef as a sea turtle glides through a school of silver fish that parts around it like a curtain. A ray lifts from the sand in a slow wingbeat of silt. The turtle rises toward the glittering surface and breaks it into the sun. Muffled underwater ambience, then one clear breath.

**That brief, planned as Veo 3.1 shots**

|  | What the shot shows | Camera | Sound |
|---|---|---|---|
| Shot 1 | A sea turtle glides through silver fish that part around it | Wide, slow tracking shot | A muffled underwater bed |
| Shot 2 | A ray lifts from the sand; the turtle rises and breaks the surface | Low angle, tilting up | The same bed, then one breath |

- One action per shot, at 4, 6 or 8 seconds; on DreamBench, Veo 3.1 clips that carry sheets run 8.
- Change the framing at each cut, keep a line inside one shot, and let an ambience run across the join.

Write each shot with the formula, or [describe the film](https://dreambench.ai/) on DreamBench, which writes, renders and cuts every shot. See [multi-shot video](https://dreambench.ai/glossary/multi-shot-video).

## Veo 3.1 prompt examples to copy

Paste these into DreamBench’s Video page or any Veo 3.1 tool.

#### A hook to camera

Vertical 9:16, 6 seconds. Handheld selfie shot. A young man in a gray beanie on a windy pier at dawn looks into the lens and says, “I tried every coffee in town. One won.” Soft cold light. Ambient noise: wind, gulls.

#### Two people talking

Widescreen 16:9, 8 seconds. Static two-shot. In a bike workshop, a customer in a wet jacket asks, “Ready by Saturday?” The mechanic spins the wheel and says, “Friday, if the parts turn up.” Warm bulb light. SFX: the freewheel ticking. Ambient noise: rain on a tin roof.

#### Product close-up: a watch

Widescreen 16:9, 6 seconds. Macro lens, slow dolly along a steel watch with a green dial on dark slate. A drop of water runs down the glass as the second hand ticks past twelve. Hard light from the left. SFX: a crisp tick.

#### Food: garlic in a wok

Vertical 9:16, 4 seconds. Static overhead shot. Sliced garlic drops into hot oil in a black wok, foams and turns gold; steam rises through window light. SFX: a loud sizzle.

#### An establishing shot

Widescreen 16:9, 8 seconds. Aerial drone shot low over a fishing village on a fjord at first light, rising to reveal snow-capped peaks. Red wooden houses, mist on the water. Photoreal, cool light warming at the peaks. Ambient noise: wind, distant engines.

#### An animal in action

Widescreen 16:9, 6 seconds. Telephoto lens, low angle. A brown hare sprints across a frosted field, swerves and stops dead, ears up. Low winter sun behind it. SFX: thudding paws. Ambient noise: a crow far off.

#### Stop motion with music

Widescreen 16:9, 8 seconds. Stop-motion claymation, static wide shot. A clay fox in a knitted scarf builds a snowman; its head falls off and the fox sighs. Warm studio light. A playful plucked-string tune begins to play. SFX: a soft plop.

#### Four shots in one clip

Widescreen 16:9, 8 seconds. [00:00-00:02] Close-up of a hand pressing a stopwatch. [00:02-00:04] Low tracking shot of spikes striking the track. SFX: spikes on rubber. [00:04-00:06] Medium shot as she leans into the line. [00:06-00:08] Wide high angle: she slows, hands on knees, the stadium empty. Ambient noise: her breath.

#### From a first frame

The woman in the picture looks up from her book, sees rain starting on the window and smiles. Slow dolly in. SFX: the first drops on the glass. Ambient noise: a quiet café.

#### Between a first and a last frame

Widescreen 16:9, 8 seconds. Static shot. A potter’s hands draw wet clay up into a tall vase as the wheel spins, then let go. Soft daylight from a high window. SFX: the hum of the wheel.

#### A two-shot sequence: the gardener

Shot 1. Widescreen 16:9, 8 seconds. Wide shot, slow pan right. A tall woman in her sixties with cropped silver hair and a green waxed jacket walks between frosted cabbages at dawn, a basket on her arm. Cold blue light. Ambient noise: boots on frozen ground, a robin.

Shot 2. Widescreen 16:9, 6 seconds. Static close-up. The tall woman in her sixties with cropped silver hair and a green waxed jacket kneels, brushes frost off a cabbage and says softly, “Just in time.” Dawn turning gold. SFX: a knife through a stalk.

> Both shots describe her in the same words and keep the dawn light; the framing changes at the cut.

## Weak Veo 3.1 prompts, rewritten

A weak prompt is usually short: every part left out is a choice Veo makes for you.

**Three weak prompts and their strong versions**

|  | Weak | Strong | What changed |
|---|---|---|---|
| A talking chef | A chef talks about his food. | Medium close-up, static. An old chef sets a bowl of soup on the pass and says, “Fifty years, the same recipe.” Warm kitchen light. Ambient noise: a busy kitchen. | A camera, a subject, a quoted line after its speaker, the sound. |
| A car advert | Epic car ad, drone then close-up then slow motion, 8K, no blur. | Widescreen 16:9, 8 seconds. Aerial tracking shot of a dark red station wagon on a coast road at sunset. Low golden light. SFX: the engine climbing. Ambient noise: surf. | Three moves became one; tags became light; “no blur” went. |
| Too much talk | Two friends talk about their trip, their jobs and next year. | Static two-shot. Two friends on a train study a map; one says, “Left at the lake?” The other laughs: “We always go left.” Ambient noise: the train. | A conversation became two lines that fit 8 seconds. |

## Common Veo 3.1 prompt mistakes and how to fix them

**Veo 3.1 prompt mistakes and fixes**

|  | What goes wrong | Fix |
|---|---|---|
| No sound written | Veo invents a soundtrack. | Name the ambience and effects every time. |
| Asking for silence | Every clip has sound. | Ask for quiet room tone; mute it in an editor. |
| A line with no speaker | A voice-over, or the wrong face. | Speaker first, then the line in quotes. |
| Too many words to say | The line is rushed or cut off. | One short line in 4 seconds, two in 8. |
| Two camera moves | Both smear. | One move; the second in another shot. |
| “No people” | A negation can bring what it names. | Write “an empty street”. |
| Describing the first frame | The words fight the picture. | Describe only what moves and is heard. |

## What a Veo 3.1 prompt can’t fix

- One prompt run twice gives two clips.
- Short lines can sound unnatural, and unwanted subtitles can appear: check every take.
- One generation stops at 8 seconds, in 16:9 or 9:16.
- A real person or product can’t be known from words; it takes a picture you have the right to use.
- On DreamBench’s Video page, a prompt past 2,600 characters is refused.

## Veo 3.1 prompts on DreamBench

- One clip: paste your prompt into the Video page and pick Veo 3.1, Fast or Lite, 4, 6 or 8 seconds, 16:9 or 9:16. It goes to Veo as written.
- One prompt: the [Veo 3.1 prompt generator](https://dreambench.ai/tools/veo-prompt-generator) writes it in this guide’s order, as text or JSON.
- A film: [describe it](https://dreambench.ai/), add up to 8 reference pictures and pick Veo 3.1. DreamBench writes and renders every shot and cuts them together; redo any shot with a note.
- Not offered: a start or end frame, a negative prompt field, or extending a clip.

The full Veo 3.1 model needs Plus or above; Fast and Lite are on every plan. The price in credits shows before you generate; the first time, create a free account and choose a plan. For prompts from words alone, see [text to video](https://dreambench.ai/glossary/text-to-video).

## Questions

- **How do you write a prompt for Veo 3.1?** To write a Veo 3.1 prompt, follow Google’s order: the camera, the subject, one action, the setting, then the style, light and mood. Add the sound in its own sentences: dialogue in quotes after its speaker, effects after “SFX:”, the background after “Ambient noise:”.
- **What is a good Veo 3.1 prompt example?** Widescreen 16:9, 6 seconds. Telephoto lens, low angle. A brown hare sprints across a frosted field, swerves and stops dead. Low winter sun. SFX: thudding paws. That is a camera, a subject, one action, light and sound.
- **How do I make Veo 3.1 talk?** Name the speaker, then put the line in quotes: A woman says, “We have to leave now.” Veo 3.1 speaks it and moves the lips in the same generation. Keep to one short line in 4 seconds, two in 8, in English.
- **Does Veo 3.1 have a negative prompt?** Google’s API has a negative prompt field: list the things, such as “wall, frame”, not “no walls”. In the main prompt, Google says to describe the result instead. DreamBench sends no negative prompt field, so describe what you want.
- **How long should a Veo 3.1 prompt be?** About 30 to 60 words for 4 seconds, 50 to 100 for 6 and 60 to 150 for 8: enough for the camera, subject, action, setting, light and sound. On DreamBench’s Video page, a Veo prompt can run to 2,600 characters.
- **How do I stop subtitles appearing in Veo 3.1 videos?** Describe a clean frame, keep lines short and leave signs and captions out. Asking for no subtitles doesn’t always work: MIT Technology Review found garbled captions even when prompts asked for none, and Google’s advice was to run the prompt again.
- **Why is the speech in my Veo 3.1 clip garbled?** Google lists natural speech in short lines as a limit it is still improving. Give each line a speaker in shot, keep it short and in English, and run the clip again if a word comes out wrong.
- **How do I write a Veo 3.1 image-to-video prompt?** From a first frame, describe what moves, not what the picture shows: the action, the camera and the sound. With reference images, write the whole shot and name the subject in words; Google takes up to three, on Veo 3.1 and Fast.
- **Can I use first and last frames with Veo 3.1 on DreamBench?** Google offers first and last frame on all three Veo 3.1 tiers; DreamBench doesn’t. Its composer draws each clip from your reference pictures, and its Video page takes words only, so write how the shot ends in the prompt.
- **How do I make a video with several scenes in Veo 3.1?** Plan one clip per moment, each a single shot of 4 to 8 seconds, and repeat each character’s description word for word. On DreamBench, describe the film: it writes and renders each shot on Veo 3.1 and cuts them together.
- **How do I keep the same character across Veo 3.1 clips?** Describe the character in the same words in every prompt and give every clip the same reference images. On DreamBench each character is drawn once as a sheet, and Veo 3.1 takes up to 3 sheets with each clip.
- **Can I paste my own prompt into Veo 3.1 on DreamBench?** Yes. DreamBench’s Video page runs Veo 3.1, Fast and Lite and sends your prompt as written, at 4, 6 or 8 seconds in 16:9 or 9:16. The first time, create a free account and choose a plan.

## Sources

- [Google Cloud: Prompting guide for Veo 3.1](https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-veo-3-1), Google Cloud: read 2026-10-31
- [Google DeepMind: Veo prompt guide](https://deepmind.google/models/veo/prompt-guide/), Google DeepMind: read 2026-10-31
- [Google Cloud: Video generation prompt guide](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/video-gen-prompt-guide), Google Cloud: read 2026-10-31
- [Google Cloud: Generate videos from text prompts](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-text), Google Cloud: read 2026-10-31
- [Google Cloud: Veo 3.1 model reference](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models/veo/3-1-generate), Google Cloud: read 2026-10-31
- [Google Cloud: Generate videos from reference images](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/use-reference-images-to-guide-video-generation), Google Cloud: read 2026-10-31
- [Google Cloud: Generate videos using first and last frames](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-first-and-last-frames), Google Cloud: read 2026-10-31
- [Google DeepMind: Veo 3.1](https://deepmind.google/models/veo/), Google DeepMind: read 2026-10-31
- [MIT Technology Review: Google’s video model has a subtitles problem](https://www.technologyreview.com/2025/07/15/1120156/googles-generative-video-model-veo-3-has-a-subtitles-problem/), MIT Technology Review: read 2026-10-31

DreamBench is an independent platform and isn’t affiliated with or endorsed by the companies named on this page.