What can PixVerse V6 do?

It has four modes: text to video; image to video with an optional last frame; reference to video, which PixVerse calls Fusion, with up to 7 named elements; and video extend. All four make clips of 1 to 15 seconds at 360p to 1080p.

Does PixVerse V6 make sound?

Yes, when you switch generate_audio on; it is off by default. The model generates ambience, effects, music and any dialogue you write in quotes together with the picture. Sound on costs more per second.

How do I name my references in the prompt?

Give each element a name of 1 to 30 characters and write @name in the prompt where it should appear, followed by a space. In our test "@trailer," with a comma right after the name was rejected, and "@trailer and" worked. Each element is a subject (the default: a person, animal, object or product) or a background (a place).

What do I get back from video extend?

One file with your source clip followed by the new seconds: a 5-second source plus 5 new seconds came back as a 10.1 second file. The source seconds are not billed. The setting and the motion carried on in our test, but the car looked different in the new part, so check the details.

When should I use PixVerse V6 Max or PixVerse V6 Effects instead?

PixVerse V6 Max is the higher-quality tier of V6 for text to video and image to video: pick it when picture quality matters more than references, extend or a last frame, and expect a higher rate per second than V6. PixVerse V6 Effects turns one photo, or two for three of the templates, into a themed clip with one of 50 ready-made PixVerse templates, with no prompt to write.

Per second of output. Resolution and the audio switch change the price per second; aspect ratio, shot type and seed do not. Video extend is billed on the new seconds. Failed or declined requests are not charged.

Are prompts and images reviewed?

PixVerse reviews prompts and images before generating; requests it declines fail and are not charged. If a request is declined, change the wording or the image and send it again.

PixVerse V6 API — Text, Image & Reference to Video

pixverse/v6/video-extend

PixVerse V6 video extend continues a clip: pick 1 to 15 new seconds and a prompt for what happens next, and get one file with your clip followed by the new part.

From $0.022/s

Run model View API

Input

Video extension

Pick the clip to continue
Picks up from here
2
What happens next
Model
Prompt146/5000
The convertible keeps following the cliff road and disappears into a short stone tunnel; the camera holds on the sea as two gulls cross the frame.
Required, 3 to 5,000 characters. Describe what happens next: the action and the camera. at least 3 characters · at most 5000 characters
Resolutionaffects price
360p, 540p, 720p (default) or 1080p. Higher resolutions cost more per second.
Durationaffects price
How many new seconds to add: whole seconds, 1 to 15 (default 5). The price is per new second; the source clip is not billed. no lower than 1 · no higher than 15
Generate audioaffects price
Off by default. On generates sound with the picture: ambience, effects, music and dialogue written in quotes. On costs more per second.

Safety checkerOff

SpicyAPI does not run a safety checker on this model. Your prompt and media go to the model as written, and the model itself decides what comes back.

Checking your account before enabling submission and pricing.

Output

Ready for a new run

Example result

Video editing

Video upscaling

Video translation

{
  "model": "pixverse/v6/video-extend",
  "input": {
    "resolution": "540p",
    "duration_seconds": 5,
    "generate_audio": false,
    "seed": 7,
    "prompt": "The convertible keeps following the cliff road and disappears into a short stone tunnel; the camera holds on the sea as two gulls cross the frame.",
    "video_url": "https://cdn.spicyapi.ai/models/examples/pixverse-v6/a0b212d7ea24b754.mp4"
  }
}

Introduction

PixVerse V6 Video extend continues a clip you already have. Send the clip and a prompt for what happens next, and choose how many new seconds to add, from 1 to 15. You get back one file with your clip followed by the new seconds. We tested extend with a PixVerse V6 clip. PixVerse reviews prompts and images before generating; requests it declines fail and are not charged.

Choose a PixVerse V6 mode by starting material

What you have at the start	Mode
a written scene	PixVerse V6 Text to video
a first frame, and optionally a last frame	PixVerse V6 Image to video
photos of people, objects or places to cast from	PixVerse V6 Reference to video
a clip you want to continue	PixVerse V6 Video extend (you are here)
a prompt or one image, and you want higher picture quality	PixVerse V6 Max
one photo and a ready-made themed template, no prompt to write	PixVerse V6 Effects

Key features & highlights

One file, source plus new seconds

In our test a 5-second source plus 5 new seconds came back as a single 10.1 second file at 540p.

The source length is not billed

Billing covers the new seconds. The seconds of the clip you send are not charged.

The setting carries on, details can drift

We extended a cliff-road clip. The new part continued the scene, with the same coastal road, sea and light, but the car looked different: its colour and shape changed. Treat extend as continuity of place and motion, not of every detail.

Sound and resolution

generate_audio is off by default and costs more per second when on. resolution sets the picture size of the new part.

Request parameters

Request parameters for PixVerse V6 · Video Extend
Parameter	Required	Description
`prompt`	Yes	Required, 3 to 5,000 characters. Describe what happens next: the action and the camera. at least 3 characters · at most 5000 characters
`video_url`	Yes	Required. The clip to continue.
`resolution`	No	360p, 540p, 720p (default) or 1080p. Higher resolutions cost more per second. `360p` · `540p` · `720p` · `1080p` · Default: `720p`
`duration_seconds`	No	How many new seconds to add: whole seconds, 1 to 15 (default 5). The price is per new second; the source clip is not billed. `≥ 1` · `≤ 15` · Default: `5`
`generate_audio`	No	Off by default. On generates sound with the picture: ambience, effects, music and dialogue written in quotes. Default: `false`
`seed`	No	Optional whole number from 0 to 2147483647. Useful to compare one change at a time; it does not guarantee identical frames. `≥ 0` · `≤ 2147483647`

Pricing

Billed on the new seconds; the source clip's length is not charged. The resolution and the audio switch change the price per second. Failed or declined requests are not charged.

Pricing
Resolution	Generate audio: No	Generate audio: Yes
`360p`	$0.0222 / second	$0.0311 / second
`540p`	$0.0311 / second	$0.04 / second
`720p`	$0.04 / second	$0.0533 / second
`1080p`	$0.08 / second	$0.1022 / second

Use cases

Finish a shot that ended too early

Describe the next beat and add a few seconds.

Build a longer scene step by step

Extend, review the result, and extend again. Each step is billed on the seconds it adds.

Hold on the scenery

Ask the camera to stay on the sea or the sky while something crosses the frame. In our test the new part showed gulls flying over the sea and ended on open water.

Explore key use-case tutorials

Prompt guide & pro tips

Describe the next beat, not the whole scene

The clip already shows who and where. Spend the prompt on what happens next and how the camera moves: in our test one sentence about the car entering a tunnel and the camera holding on the sea was enough. Keep the resolution and the sound setting the same as the source clip so the join does not change look or audio, and ask for only as many new seconds as the next beat needs, because the new seconds are what you pay for.

Check the details after the join

Because the car changed colour and shape in our test, look for props, vehicles and wardrobe that must stay the same, and re-run with a changed prompt or seed if they drift.

Compare with a seed

seed keeps runs comparable while you change one thing at a time. It does not guarantee identical frames.

API quick start

# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $SPICY_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
  -d '{
  "model": "pixverse/v6/video-extend",
  "input": {
    "duration_seconds": 5,
    "generate_audio": false,
    "prompt": "A slow aerial shot drifting over a misty pine forest at sunrise",
    "resolution": "720p",
    "video_url": "https://example.com/reference.mp4"
  }
}'

# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $SPICY_API_KEY"

See the API reference for authentication, the complete schema, polling, and error handling.

Missing the model you need?

Tell us what to add and where you would use it. Requests go to the people who decide what gets onboarded next.

PIXVERSE'S FOUR-MODE VIDEO MODEL

PixVerse V6 API at a glance

Text, first and last frames, up to 7 named references and clip extension, in 1 to 15 second clips from 360p to 1080p, with optional sound.

Clip length: 1-15 S; Any whole number of seconds on every endpoint, billed per second.
Up to 1080p: 1080P; 360p, 540p, 720p (default) or 1080p; higher resolutions cost more per second.
Modes: 4; Text, image with an optional last frame, reference with @names, and extend.
Named references: 7; Up to 7 subjects or backgrounds, each called by name in the prompt with @name.
Optional audio: SOUND; Switch generate_audio on for ambience, effects, music and quoted dialogue; it costs more per second.
Model time: ~40 S; A 5-second 540p clip took 37 to 52 seconds of model time in our runs; the extend took 67.

Results

What you can make with PixVerse V6

From text
A close-up that rises to an aerial
One paragraph opened on a close shot of the laughing driver with her scarf flying, then the camera rose to a wide aerial of the teal convertible on the cliff road above the ocean. The prompt asked for the drone to pull back and rise; the cut from close-up to aerial came from the model.
Text to video, 5 seconds at 540p, 16:9, no audio.
From frames
Pin the start and the end
A first frame of a woman in a rust coat at the open slit of an observatory dome and a last frame of her leaning on the railing became one move: she turns, walks along the railing, leans back and crosses her arms, ending on the last frame.
Image to video with a first and a last frame, 5 seconds at 540p.
From references
Cast a scene by name
A photo named maya and a photo named trailer, called as @maya and @trailer in the prompt, gave a dusk shot of her walking toward the camera with an enamel mug. Her face, hair and jacket match the reference, and so does the trailer. One frame shows a second figure briefly in the doorway.
Reference to video from two elements, @maya and @trailer, 5 seconds at 540p.
From a clip
Carry a scene on
The 5-second cliff-road clip plus 5 new seconds came back as one 10.1 second file. The coastal road, the sea and the light carried over; the car looked different in the new part, with another colour and shape.
Video extend: the cliff-road clip plus 5 new seconds, one 10.1 second file at 540p.

Content test

Is PixVerse V6 uncensored? What we tested

We ran each level on our live setup and checked every clip frame by frame. A pass means the content was actually on screen, not just that the request went through.

Content levels tested on PixVerse V6, by mode
Level	Text to video
MildThe lightest prompts in the test	Not tested yet
ModerateWhere filtered models start to refuse	RefusedTested Oct 25, 2026
StrongThe middle of the scale	Not tested yet
Very strongWhere most models start to hold back	Not tested yet
MaximumThe top of the scale	Not tested yet

Renders: on screen as written. Softened: generated, but milder than what was written. Refused: rejected. Not tested yet: we have not run that level on that mode.

How each mode behaves

All modes: Clips run 1 to 15 seconds at 360p to 1080p. PixVerse reviews prompts and images before generating; requests it declines fail and are not charged.
Image to video: The clip follows the aspect ratio of your first image; there is no aspect_ratio field. shot_type multi cannot be combined with a last frame. Images are JPG, JPEG, PNG or WebP, up to 20 MB.
Reference to video: Send 1 to 7 elements and put a space after each @name: in our test "@trailer," with a comma right after the name was rejected because the name did not match. There is no multi-shot switch on this endpoint, and small artifacts can appear, such as a second figure in one frame of our sample.
Video extension: The new part continues the setting and the motion, but details can drift: in our test the car changed colour and shape. We tested extend with a PixVerse V6 clip.

Use cases

What to build with the PixVerse V6 API

Opening and establishing shots
Write the place, the light, one camera move and the look. Camera moves you write in the prompt are followed, and sound can come with the picture.
Stills that move between two poses
Give a first frame and a last frame and let the model fill in the move between them.
Consistent people, products and places
Send up to 7 named elements, subjects or backgrounds, and cast them into a new scene with @name in the prompt.
Vertical and ultrawide frames
Text and reference to video take 9:16, 3:4, 2:3, 1:1, 4:3, 3:2, 16:9 and 21:9. The price per second is the same for every aspect ratio.
Longer scenes in steps
Extend a clip by 1 to 15 seconds at a time, billed on the new seconds. Check props and vehicles after each join, because details can drift.

Prompts

PixVerse V6 prompt examples you can copy

Paste one into the playground as it is, then change one thing at a time.

Text to video
Place, action, camera, look
Result from this prompt
Golden hour on a cliffside coastal highway. A vintage teal convertible rounds a bend, a woman in a white silk headscarf and round sunglasses at the wheel, laughing as the scarf ripples behind her. The drone pulls back and rises to reveal the Pacific and the road winding into the haze. Warm 35mm film look.
Why it works. The prompt gives the place and light, the action, one camera move and the film look in the order they happen, so the model has a clear sequence to follow.
Try this prompt
Image to video
Describe the next beats, not the picture
Result from this prompt
The sodium lamp above the pumps flickers fully on. He folds the road map, slides it into his jacket and walks to the driver door as a truck passes, its headlights sweeping across him. Handheld, slow push-in.
Why it works. The first frame already shows the man, the car and the station, so the prompt spends its words on what he does next and how the camera moves.
This mode starts from your own image: add it in the playground before you run.
Try this prompt
Image to video
Pin the ending with a last frame
Result from this prompt
She takes one last look through the open slit of the dome, then turns around, leans back against the steel railing and crosses her arms with a small satisfied smile. Static wide shot.
Why it works. With a last frame set, the prompt just has to describe how she gets from one pose to the other, plus the camera.
This mode starts from your own image: add it in the playground before you run.
Try this prompt
Reference to video
Call each element with @name and a space
Result from this prompt
At blue hour in the high desert, @maya steps out of the door of @trailer and stretches, then walks toward the camera holding a steaming enamel mug as a string of warm bulbs flickers on along the awning. Slow dolly back.
Why it works. Each @name matches an element you sent and has a space after it; the rest of the prompt describes the action and the camera.
Try this prompt
Video extension
Say what happens next
Result from this prompt
The convertible keeps following the cliff road and disappears into a short stone tunnel; the camera holds on the sea as two gulls cross the frame.
Why it works. The clip already shows the road and the sea, so the prompt describes the next beat and where the camera goes.
Try this prompt

Compare

PixVerse V6 vs similar models: which one to pick

Model	Modes	From	Content	Pick it when
PixVerse V6This page	Image to video · Reference to video · Text to video · Video extension	$0.022 / output second	Filtered — This model refuses requests beyond its own limits outright. Measured, not assumed — models we have not measured carry no label at all.	You want text, first and last frames, @name references and extend in one family, with a multi-shot switch on text and image to video.
PixVerse V6 Max	Image to video · Text to video	$0.027 / output second	Light — Content restrictions are lighter, but the model may still limit some text-only prompts. Image-guided modes can offer more control. Applicable content limits still apply.	You want higher picture quality from a text prompt or a single image and accept a higher rate per second; it has no references, no extend and no last frame.
Kling 4.0	Image to video · Reference to video · Text to video · Video editing · Video extension	$0.113 / output second	Filtered — This model refuses requests beyond its own limits outright. Measured, not assumed — models we have not measured carry no label at all.	You need clips longer than 15 seconds, which PixVerse V6 does not make.
Seedance 2.0	Image to video · Reference to video · Text to video · Video extension	$0.07 / output second	Uncensored — This model declares no content filter. SpicyAPI forwards the request without adding a platform content scan; legal and consent duties remain with the customer.	You want to run the same brief through another publisher's video model and compare the results; keep the length and aspect ratio the same.

Pricing

PixVerse V6 API pricing

A 5-second clip starts at $0.111 at 360p and $0.4 at 1080p.

The playground shows the exact price for your settings before you run. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.

See every price tier

Get started

How to use the PixVerse V6 API in three steps

Step 1
Draft small
Find the framing with a short clip at 540p or lower, then render the version you keep at 720p or 1080p. Resolution changes the price per second.
Step 2
Write the camera move
Name the move in the prompt, such as "the drone pulls back and rises" or "slow dolly back". Camera moves you write are followed.
Step 3
Describe the sound when audio is on
With generate_audio on, name the ambience, effects or music you want and put any dialogue in quotes.

Open the playground Get an API key

Parameter reference and code samples are in the API and Readme tabs at the top of this page.

Why SpicyAPI

Why run PixVerse V6 on SpicyAPI

No filter from us
We add no content filter of our own. What the model renders is what you get back.
Tested, not guessed
Content labels come from real generations we ran and checked, with the test date on the page.
Pay per result, in dollars
No subscription and no credits. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.
One key for every model
One API key and balance cover image, video, audio and chat models, so you can explore more models from the same account.

FAQ

PixVerse V6 API FAQ

About PixVerse V6

PixVerse V6 brings PixVerse's video model to one API: text to video, image to video with an optional last frame, reference to video with up to 7 named elements, and video extend. Clips run 1 to 15 seconds at 360p to 1080p, with optional generated sound and eight aspect ratios on text and reference to video. PixVerse reviews prompts and images before generating, and declined requests are not charged. SpicyAPI bills per second of output in US dollars, with no subscription.

All PixVerse models2

PixVerse V6 MaxNewText to video · Image to videoLight — Content restrictions are lighter, but the model may still limit some text-only prompts. Image-guided modes can offer more control. Applicable content limits still apply.From $0.027per output second
PixVerse V6CurrentText to video · Image to video · +2From $0.022per output second

Workbench

Input

Pick the clip to continue

What happens next

Output

Introduction

Choose a PixVerse V6 mode by starting material

Key features & highlights

One file, source plus new seconds

The source length is not billed

The setting carries on, details can drift

Sound and resolution

Request parameters

Pricing

Use cases

Finish a shot that ended too early

Build a longer scene step by step

Hold on the scenery

Prompt guide & pro tips

Describe the next beat, not the whole scene

Check the details after the join

Compare with a seed

API quick start

A close-up that rises to an aerial

Pin the start and the end

Cast a scene by name

Carry a scene on

How each mode behaves

Opening and establishing shots

Stills that move between two poses

Consistent people, products and places

Vertical and ultrawide frames

Longer scenes in steps

Place, action, camera, look

Describe the next beats, not the picture

Pin the ending with a last frame

Call each element with @name and a space

Say what happens next

Draft small

Write the camera move

Describe the sound when audio is on

No filter from us

Tested, not guessed

Pay per result, in dollars

One key for every model

What can PixVerse V6 do?+

Does PixVerse V6 make sound?+

How do I name my references in the prompt?+

What do I get back from video extend?+

When should I use PixVerse V6 Max or PixVerse V6 Effects instead?+

How is it priced?+

Are prompts and images reviewed?+

About PixVerse V6

All PixVerse models2

Related models and next tasks