Pricing
VIDEONative audioGoogle

Veo 3.1Picture and sound, one render.

You stop scoring silent drafts. Veo 3.1 renders picture and audio in one pass. Both tracks, one take.

See the takes
  • 4 · 6 · 8s
  • 720p · 1080p
  • Native audio
  • Up to 3 references
  • 16:9 · 9:16

Free account · ⚡50 on signup · No card

● Shot on Veo 3.1

Both tracks, one pass

Documentary, claymation, macro, anime. Six takes off two variants, each with the prompt that produced it.

Motorcycle on the salt flat8s

Motorcycle on the salt flat

Veo 3.1 · 8s

“Make a dreamy retro commercial video of a vintage motorcycle alone on a mirrored salt flat.”

Clay office worker8s

Clay office worker

Veo 3.1 Fast · 8s

“Make a handmade claymation video of a clay office worker collapsing under a pile of paperwork, then popping up with a "Ta-da!" sign.”

Pour-over, close4s

Pour-over, close

Veo 3.1 · 4s

“Make a cinematic video of a steaming coffee cup.”

Car macro

Car macro

Veo 3.1 Fast

“Make cinematic showcase of a car, with macro shots.”

Into the sea cave

Into the sea cave

Veo 3.1

“Make a cinematic video of ocean surface, then show underwater.”

Retro CRT pitchman8s

Retro CRT pitchman

Veo 3.1 Fast · 8s

“Make a retro infomercial video of an 80s pitchman presenting to camera, shown on a vintage CRT screen.”

● Release notes

What's new in Veo 3.1

01

Chapter 01 · Sound

Audio in the same take

Richer audio, more narrative control, more realism. Prompt the sound as part of the shot.

The call from the pass
02

Chapter 02 · The brief

Tighter prompt adherence

Veo 3.1 sharpens adherence over the previous generation. Write the frame you want, read it back.

The gold-leaf sign painter
03

Chapter 03 · The finish

1080p on the long stop

1080p renders at 8 seconds, for ⚡652. Shorter lengths deliver 720p.

A city cut from card
04

Chapter 04 · The cast

References that hold

Multiple reference images control characters, objects and style. Veo 3.1 takes up to 3.

The coat, all the way round

● People, on camera

A face, a voice, one render

Two creator selfies and a bookseller mid-sentence. The dialogue is in the take, not added after.

The bookshop take ships the prompt it was rendered from. The two creator takes carry recreate prompts.

Creator at the desk8s

Creator at the desk

Veo 3.1 · 8s

“A woman in her late twenties talks straight into a handheld phone camera at a sunlit home desk, holding a small unbranded pack up to the lens mid-sentence. Selfie framing, window light, a plant and a mug behind her. Audio: her voice close on the mic, a quiet room, faint street noise.”

Walk and talk, golden hour8s

Walk and talk, golden hour

Veo 3.1 · 8s

“A woman in a yellow top walks a pastel side street at golden hour, filming herself at arm's length, laughing mid-sentence. Handheld selfie framing, low sun flaring behind her, parked cars and shuttered shopfronts. Audio: her voice, footsteps, a scooter passing.”

The bookseller picks one8s

The bookseller picks one

Veo 3.1 · 8s

“A bookseller in a narrow shop turns a paperback to camera and says, one single time and never repeating herself, "This is the one I keep buying for people who say they don't read." She then smiles and looks down at the cover in silence. Handheld, dust in window light, shelves behind. Audio: her voice speaking that line exactly once, a page turn, quiet street noise through glass. No repeated words, no looped dialogue.”

Made with Veo 3.1 on OpenClips

Verbatim prompt
A long passenger train curves along a single track through a deep wooded valley at golden hour, low sun flaring between tall pines, wildflowers and long grass thrashing in the draught as the carriages pass. The camera tracks alongside at the speed of the train, the forest and a hazy ridge rising above it. Audio: steel wheels ringing on rail, a low horn far off down the valley, wind through grass.
Remix in composer →

Generated on OpenClips · veo-3.1 · September 2026

Verbatim prompt
Floodlights snap on one bank at a time over an empty football stadium at dusk, rain drifting through the beams, camera tilting up from the centre circle. Audio: four heavy contactor thuds, rain on plastic seats, a low wind through the stands.
Remix in composer →

Generated on OpenClips · veo-3.1 · September 2026

Verbatim prompt
A young woman stands inside a phone box at night, filmed from outside through rain-streaked glass, mouthing words into the handset. We never hear her voice. Heavy rain falls through the streetlight above, running down the panes and softening her face, wet road glinting in the foreground. The camera holds completely still. Audio: rain hammering on the roof of the box, water running in a gutter, one car passing far off. No speech, no music.
Remix in composer →

Generated on OpenClips · veo-3.1 · September 2026

● How to

How to use Veo 3.1 on OpenClips

Three steps from a blank box to a take with sound. The page works logged out. Sign-in gates the render only.

  1. Veo 3.1 · Composer

    Press a themeIt loads a real prompt

    “Make a dreamy retro commercial video of a vintage motorcycle alone on a mirrored salt flat.”

    “Make a handmade claymation video of a clay office worker collapsing under a pile of paperwork, then popping up with a "Ta-da!" sign.”

    “Make a cinematic video of a steaming coffee cup.”

    “Make a retro infomercial video of an 80s pitchman presenting to camera, shown on a vintage CRT screen.”

    01 ·

    ≈ 2 min

    Pick a theme or write the shot

    Press a theme to load the prompt that made that take. Or describe the frame, the motion and the sound.

  2. Veo 3.1 · Composer
    16:99:168s

    02 ·

    ≈ 1 min

    Attach references, set the frame

    Up to 3 stills hold a face, a product or a palette. Pick the ratio and the length.

  3. Veo 3.1 · Composer
    Rendering
    From ⚡326Generate

    03 ·

    ≈ 4 min

    Read the price, then render

    The exact ⚡ shows on the button. Takes land in your library.

Render your first take

● Two ways in

The porch at first light

Veo 3.1 text to video

Write the shot, the move and the sound. One paragraph comes back as an eight-second take. Steal working setups from the community wall.

Flowers on the kitchen table

Veo 3.1 image to video

Start from the frame you already have. Up to 3 stills hold a face, a product or a palette. Full rates live on the pricing page.

Generate with Veo 3.1 →

● Press play

The sound is already in the file

Three takes, three kinds of audio. Dialogue, ambience and one effect. Nothing was scored after. The dialogue tile carries a recreate prompt.

The office testimonial
Dialogue

A line, on the picture

Verbatim prompt
A woman in a beige blazer sits at a table in a glass-walled office and speaks directly to camera, a white cup in front of her. Medium shot, soft daylight, an empty meeting room behind. Audio: her voice, room tone, a distant keyboard.
Wheat under the storm
Ambience

Wind, thunder, first drops

Verbatim prompt
Wide shot of a wheat field under a breaking thunderstorm, the crop bending in waves as the light drops. Slow push-in, no people. Audio: wind through wheat, distant thunder, first heavy drops.
One hiss, then steam
Effects

One hiss, then steam

Verbatim prompt
Close-up of a blacksmith plunging glowing steel into a quench tank, steam bursting up around the tongs. Dark forge, orange rim light, slow motion. Audio: one violent hiss, steam, a hammer set down.
Generate with Veo 3.1 →

● Use cases

What the eight seconds are for

One take is a hook, a hero shot or a demo. Here is the same model on five briefs.

1080p · 8s
Automotive

The hero shot on the coast road

Camera direction in words. The chase move and the sun both land.

Native audio
Product film

A scan you did not composite

The laser pass and its sound come back in the same render.

16:9
Food

Top down, softly turning

Tabletop light and a slow move. No rig, no re-shoot.

9:16
Lifestyle · 9:16

A vertical cut for the feed

Portrait comes out of the composer. It is a setting, not a crop.

1080p · 8s
Genre film

A set you did not build

Haze, practicals, a long corridor. Production design in one sentence.

Generate with Veo 3.1 →

● Every format

Veo 3.1 video examples

A full cafe, an aquarium, a cello under concrete. Press any tile for sound.

Five of these ship the prompt they were rendered from. The night drive carries a recreate prompt.

One line across a full room8s

One line across a full room

Veo 3.1 · 8s

Verbatim prompt
A waiter crosses a packed evening cafe carrying two plates and calls out one line, "Behind you, coming through," exactly once and never repeating himself. Handheld, warm lamps, diners blurred around him. Audio: his voice cutting through the room, crowd chatter, cutlery, a coffee machine.
Tea set, cinematic4s

Tea set, cinematic

Veo 3.1 Fast · 4s

Verbatim prompt
Make a cinematic video of a tea set.
Neon traffic, wipers on8s

Neon traffic, wipers on

Veo 3.1 · 8s

Verbatim prompt
A woman drives through a neon city in heavy rain, first looking out through the streaked windshield at red tail lights, then a profile shot of her at the wheel. Handheld, cold neon on wet glass. Audio: rain on the roof, wipers, traffic and a low engine.
Moon jellyfish, backlit8s

Moon jellyfish, backlit

Veo 3.1 · 8s

Verbatim prompt
Macro of moon jellyfish drifting in a dark aquarium, backlit so their bells glow. Slow lateral drift, deep blue. Audio: muffled water, a low hum.
One bow stroke, under concrete8s

One bow stroke, under concrete

Veo 3.1 · 8s

Verbatim prompt
Anamorphic 40mm, shallow depth of field, shot on Kodak 5219 with a warm grade and visible grain. A lone cellist under a concrete overpass at night, lit only by a single sodium lamp above and behind him, smoke and dust hanging in the beam. The camera pushes in eight inches over the whole shot, nothing else moves. Audio: one long low bow stroke, hard echo off the concrete, a single car passing far away.
Closed, but sit down8s

Closed, but sit down

Veo 3.1 · 8s

Verbatim prompt
Anamorphic, warm grade, deep blacks. A bartender stands under a single low hanging lamp in an otherwise dark empty bar, bottles catching one edge of light behind him, everything else falling to black. He looks up and says, one single time and without repeating himself, "We are closed, but sit down." No brand names or logos anywhere in frame. The camera holds completely still. Audio: his voice, a glass set on wood, a fridge hum.

● Nine by sixteen

Cut for the feed

Veo 3.1 renders 9:16 as readily as 16:9, audio included. Set the frame in the composer. The take comes back cut for a phone.

The skate take ships its own prompt. The lantern and street takes carry recreate prompts.

The letter by lantern8s

The letter by lantern

Veo 3.1 · 8s

“An older woman sits at a wooden table and reads a letter by the light of an oil lantern, rain running down the window behind her. Vertical frame, warm lantern key against cold blue outside, slow push-in. Audio: rain on glass, paper moving, a quiet breath.”

Lantern street, after rain6s

Lantern street, after rain

Veo 3.1 · 6s

“A lantern-lit food street after rain, steam rising off the grills, shoppers moving over wet cobbles under rows of red lanterns. Vertical frame, camera pushing slowly down the middle of the lane. Audio: crowd chatter, sizzling, a cleaver on a board.”

Drop in, land it8s

Drop in, land it

Veo 3.1 · 8s

“A skateboarder drops into a concrete bowl and lands a trick, the camera following from the coping. Vertical frame, late afternoon sun, graffiti walls. Audio: wheels on concrete, the pop of the board, one shout.”

Open the full reel

● Four things it does

What you are actually buying

One line, one lamp
Audio

Picture and sound, one pass

Prompt the room and the line together. Veo 3.1 writes both. Short pieces skip the sound design pass.

Verbatim prompt
Anamorphic, warm grade, deep blacks. A voice actor stands at a studio microphone under one hard overhead lamp, everything around him falling to pure black, dust visible in the beam. He leans in and says, one single time and without repeating himself, "Take it from the top." The camera holds completely still. Audio: his voice close on the mic, headphone bleed, room tone.
Through the kitchen, uncut
Direction

Say the move, read it back

Name the move and the light. The camera crosses a whole kitchen without a cut.

Verbatim prompt
One continuous take through a busy restaurant kitchen: the camera moves past a grill, around a prep table and out to the pass, never cutting. Handheld, tungsten and steam. Audio: sizzle, knives on boards, short calls between cooks.
Dolly to the notebook
Identity

Up to three stills hold the subject

One still per face, product or palette. Identity holds across the take and across re-rolls.

A face to hold
A product to hold
A palette to hold
Verbatim prompt
Slow dolly-in through a doorway into a lamp-lit study, ending on an open notebook on the desk. Locked horizon, shallow focus, dust in the beam. Audio: floorboard creak, a clock, rain outside.
Studio portrait, teal
Framing

Nine by sixteen, natively

Pick the ratio before you render. The take arrives cut for a phone. This prompt is written to recreate the take.

Verbatim prompt
A close portrait of a bearded man in a yellow shirt against a deep teal backdrop, turning his head slowly toward the light. Vertical frame, one hard key from the side, skin detail held. Audio: room tone, one breath.

● One take

Two voices and the weather

A line each, rain on the umbrella, a bus pulling in. All of it in one render.

Veo 3.1 · 8s · 1080p · native audio
Verbatim prompt
A father and his young daughter wait at a bus stop in light rain. She asks "Is it late again?" and he laughs and says "Always." Two-shot, overcast, shared umbrella. Audio: both voices, rain on the umbrella, a bus pulling in.

● The sheet

Veo 3.1 and Veo 3.1 Fast

Same model, two speeds. Fast trades a few minutes of render time. Roughly a third of the price.

SpecVeo 3.1Veo 3.1 Fast
StatusLive in the composerLive
Resolutions720p · 1080p720p · 1080p
Clip length4s · 6s · 8s4s · 6s · 8s
1080p noteRenders at 8s onlyRenders at 8s only
AudioNative audioNative audio
Reference imagesUp to 3Up to 3
Aspect ratios16:9 · 9:1616:9 · 9:16
Render speed~4–6 min per clip~2–3 min per clip
From⚡326⚡123

● Spec sheet

Veo 3.1 model parameters

Modality
video
Resolutions
720p, 1080p
Aspect ratios
16:9, 9:16
Duration
Up to 8s
Audio
Native audio

Figures reflect OpenClips defaults, August 2026.

Compare every model that renders sound.Compare every model that takes several references.

● One model, ten disciplines

What people point it at

Documentary, claymation, macro and anime came off the same composer.

Documentary portraitProduct macroAutomotiveClaymationCel-shaded animeRetro infomercialUnderwaterDialogue scenesAmbience and effectsVertical cutdowns

● Quoted first

Priced per render. Quoted first.

Most used

Veo 3.1

4s · veo-3.1

326

$3.26 at pack rate

The finishing lane. 720p at the short stop.

Veo 3.1 Fast

4s · veo-3.1-fast

123

$1.23 at pack rate

Draft lane. 1080p at 8s is ⚡245.

1080p finish

8s · veo-3.1

652

$6.52 at pack rate

The delivery cell. 1080p renders at 8 seconds.

● The price grid

Every length, every tier

Read the cell you plan to render.

Veo 3.1

Length720p1080p
4s⚡326$3.26-
6s⚡489$4.89-
8s⚡652$6.52⚡652$6.52

Veo 3.1 Fast

Length720p1080p
4s⚡123$1.23-
6s⚡184$1.84-
8s⚡245$2.45⚡245$2.45

Dollars are the pack rate: $10/1,000⚡. Plans buy credits cheaper. The exact ⚡ price shows on the Generate button before you spend.

● OpenClips MCP

Drive Veo 3.1 from your AI agent

No console and no syntax. Your agent briefs Veo 3.1 through the OpenClips MCP server. It queues the render and hands back the take.

Server addresshttps://mcp.openclips.ai
ClaudeClaude CodeChatGPTCursorVS CodeGrokHermes agentOpenClawOther MCP client

  1. Step 01

    Say it in plain words

    One sentence is the brief. Subject, sound, platform. No syntax to learn.

  2. Step 02

    Your agent writes the shot

    Frame, length, references and the ⚡ quote. Drafted and queued for you.

  3. Step 03

    The take lands in your library

    Download it, remix it, or ship it. Credits settle automatically.

Explore the MCP server

● Director's notes

Tips for best results with Veo 3.1

Four habits that separate a clean take from an expensive re-roll.

  • Prompt the sound, not just the shot

    Name the ambience and the one line of dialogue. Veo writes audio with the picture.

    Rain on glass, one whisper

  • Direct the camera in words

    Say the lens move and the light. Veo follows camera direction closely.

    Slow push, golden hour

  • Reference every recurring subject

    One still per face or product, up to three. Identity holds across the take.

    One still per subject

  • Draft on Fast, finish on 3.1

    Test the beat at ⚡123. Move the keeper across for the finish.

    Same composer, same library

● From the community

Made on Veo, on OpenClips

Real renders from real accounts. Open one and remix the setup.

The Letter

The Letter

“A grandmother reads to her grandson, warm lamp light, gentle push-in, soft rain outside”

OC StudioVeo 3.1

Bottle Hero Loop

Bottle Hero Loop

“Slow-motion hero shot of a matte-black bottle rotating on wet stone, studio rim light”

OC StudioVeo 3.1

The PAS Ad

The PAS Ad

“Problem-agitate-solution ad: messy desk, panic, then calm — one continuous shot”

OC StudioVeo 3.1

See the community wall →

● From the blog

Read the director's cut

Model Guides

Picking your A-camera

Three flagship video models, three personalities. Which one to roll.

Read it →
Prompt Craft

The video prompt formula

Six beats that turn one line of text into a trailer-grade clip.

Read it →
Prompt Craft

The aspect ratio cheat sheet

Which frame wins where, and how to compose for it before you render.

Read it →

● The gaps

What can Veo 3.1 do that OpenClips cannot?

The vendor ships these. The composer does not.

  • Google ships Extend for Veo, for shots a minute or more long. The composer has no field for it.

    Read the vendor page

  • Google ships Frames to Video, bridging a start image and an end image. The composer has no field for it.

    Read the vendor page

  • Google ships Insert, which adds an object into a finished Veo shot. The composer has no field for it.

    Read the vendor page

● Questions

Frequently asked questions

Veo 3.1 is free to try. A new OpenClips account gets ⚡50 on signup with no card. Video renders draw from that wallet, and a Veo 3.1 clip costs more than the grant, so top up when the deliverable matters. Packs start at $10/1,000⚡.

Veo 3.1 is Google DeepMind's video model, built for cinematic shots with sound baked in. Google DeepMind describes Veo under the tagline "Video, meet audio." On OpenClips you reach it through the same composer as every other model, on one credit wallet and one library. No separate vendor subscription, no export dance between tools, no waiting list.

Veo 3.1 Fast is the quicker, cheaper lane: from ⚡123 at ~2–3 min per clip, against ~4–6 min on Veo 3.1 from ⚡326. Both cover the same 4/6/8-second lengths, both reach 1080p, both carry native audio. Draft on Fast, finish on Veo 3.1.

Yes, Veo writes the sound with the picture. Google DeepMind says Veo offers new levels of control, consistency, and creativity, now across audio. Prompt the sound the way you prompt the shot: rain on glass, muffled traffic, one whispered line. It comes back in sync, so short pieces skip the sound design pass entirely.

A render is 4, 6 or 8 seconds, picked in the composer, and the price follows the length: ⚡326 at 4 seconds, ⚡489 at 6, ⚡652 at 8. Eight seconds is the ceiling per take on OpenClips today. Need more screen time? Render a few takes on the same references and cut them together.

Yes. 1080p is the 8-second output on both Veo 3.1 and Veo 3.1 Fast. Shorter lengths (4s and 6s) render at 720p. Pick 8s in the composer when the deliverable needs the higher resolution.

Up to 3 reference images per render on Veo 3.1 and Veo 3.1 Fast. Use them to hold a character, a product or a palette across takes. Attach them in the composer alongside the prompt.

Veo can be used in Gemini and in Google Flow, and developers can build with it through the Gemini API video docs. OpenClips is the consumer route to it: one login, one credit wallet, and every other video model sitting beside it. No stack of per-vendor plans. Start free, pay per render, keep every clip in one library.

Yes. Veo 3.1 clips rendered on OpenClips are yours to use commercially. Google's own Veo terms cover the model, and OpenClips serves it under the same license. Keep the render receipt (the library entry) for provenance.

● The receipts

Where do these Veo 3.1 facts come from?

Every link goes to the vendor's own page.

Your shot. With sound.

One render, both tracks, the price on the button.

Generate with Veo 3.1

Or roll with a sibling model

Seedance 2.5 ›Nano Banana Pro ›Kling 3.0 ›All models ›

Veo 3.1 in one pass

Prompting sound as part of the shot changes the draft you get back. Instead of a silent plate you have to score later, the take arrives with the room in it.

Two lanes, one wallet

Veo 3.1 Fast renders in ~2–3 min from ⚡123. Move the keeper to Veo 3.1 for the finish. Same composer, same library, no re-uploads between the two.

Where the take stops

One render is 4, 6 or 8 seconds. 1080p is the 8-second output. Longer pieces are a cut of several takes, and references are what hold a face or a product steady across them.

● Where to go next

Related models

video

Seedance

Flagship

ByteDance · video

video

Hailuo

MiniMax · video

video

Kling

Hot

Kuaishou · video

video

Runway

Runway · video