Skip to main content
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.

Vidu Q3 Review: Literal Motion, Priced by Resolution

Pick Vidu Q3 when you need 1080p and fine control over how much a still moves; for the same family at a lower price, look at Q3 Turbo or Q3 Pro instead.

Open the Playground

Spicy Index

74.0/ 100

Preliminary

Freedom Score

93.3/ 100

Content levels

Tested 6 of 6 probe types · 19 runs

  • Mild · The lightest prompts in the test

  • Moderate · Where filtered models start to refuse

  • Strong · Mid-scale; must be tested before a model is ranked

  • Very strong · Where most models start to hold back

  • Maximum · The top of the scale, scored in two halves: class A and class B

Price
$0.75 per 5-second clip
Median time
82 s
Success rate
Collecting data

Vidu Q3 does one job: it takes a still image and a prompt describing the movement that follows, and animates 1 to 16 seconds of video at 540p, 720p or 1080p. It's the one tier in the Q3 line with a movement_amplitude control (auto, small, medium or large), and it's the one that goes past 720p; it's priced by resolution rather than a flat rate: $0.07 a second at 540p, $0.15 at 720p, $0.16 at 1080p.

What Vidu Q3 made in our tests

Straight from our test runs. Copy a prompt or open it in the playground to try it yourself.

  • Image to video

    The wind gusts and the long grass runs in waves across the clifftop; her white dress and hair stream sideways and snap. She turns from the sea into the wind toward the camera, catching the hair off her face with one hand and laughing, then spreads both arms and leans into the gust. Gulls cross far below. Slow steadicam arc around her, bright overcast daylight, cool green and pale blue, sea roar and wind.

    Try this model

  • Image to video

    She stretches both arms up over her head so the oversized shirt lifts and slides off one shoulder, then lets herself fall slowly backwards across the white bed, bare legs drawing up and her bare feet crossing at the ankles in the foreground. She turns her head on the duvet toward the camera, pushes the messy hair off her face and smiles into the lens. Sheer curtain breathing in the sunlight, dust in the beam, high-key morning light, slow gentle handheld drift.

    Try this model

Make your own with Vidu Q3

Open the Playground

The verdict

Vidu Q3 is one of three tiers we tested from this family — the others are Q3 Pro and Q3 Turbo, reviewed separately — and in our 27 September side-by-side comparison against a dozen other image-to-video families, all three behaved like the same model tuned three ways: careful with camera motion, faithful to what the source frame is wearing, and consistent about not adding a scene you didn't ask for. Q3 itself is the one to reach for if you need 1080p output or want to dial motion up or down directly rather than trust the model's default.

Choose Q3 when you need 1080p or direct movement-amplitude control. Compare costs at the same resolution: Turbo is 40% cheaper at 540p, while Pro is fixed at 540p. Audio is available across the Q3 line.

How it scores. Freedom Score comes from the existing content tests. See the test report for the evidence and current scores.

Try Vidu Q3 →

Key facts

  • Pass rate: across 19 content-ladder runs between 14 and 25 September 2026, every rung rendered as asked except L4 Intense, where one of three tries rendered and two came back softened.
  • Price: $0.07/s at 540p, $0.15/s at the default 720p, and $0.16/s at 1080p. A default 5-second clip costs $0.75. Check the model page estimate for your selected settings.
  • Limits: 1 to 16 seconds, image-to-video only — there's no text-to-video or reference-to-video endpoint on this family. movement_amplitude (auto/small/medium/large) is unique to this tier among the three Q3 versions.
  • Sources: every scored run is in the Vidu Q3 test report, the scoring rules are on our review methodology page, and the full parameter list is on the Vidu Q3 model page.

What we liked and what held it back

What we liked

  • It goes to 1080p; the other two Q3 tiers don't. Q3 Pro is fixed at 540p and Q3 Turbo tops out at 720p, so this is the family's route to a full-resolution deliverable.
  • movement_amplitude gives you a lever the siblings lack. Neither Q3 Pro nor Q3 Turbo exposes a motion-strength control; here you can ask for small on a near-still portrait or large on an action shot.
  • A locked-off camera reads as locked off. On a cat leaping onto a bookshelf, the frame held steady through the leap and the landing kept the shelf at the same angle as the source photo — no sudden perspective shift, which we did see from Q3 Pro and Q3 Turbo on this exact prompt.
  • It doesn't add clothing changes you didn't write. On a woman turning from a window toward the camera in a silk dress, it kept the dress on and followed the described turn and hair movement, matching what we saw from Q3 Pro and Q3 Turbo on the same still.
  • Fine bubble and steam detail on a pour-over shot. Water pouring into a dripper, foam forming, steam rising — the physics-heavy prompt came back with the same convincing bloom we saw from the other two tiers, suggesting this quality baseline is shared across the whole Q3 line rather than something you pay extra for.
  • A genuine, understated smile. Asked for a woman who "slowly breaks into a genuine smile," Q3's version stayed closer to the word "genuine" than the wide, toothy grins several other families gave the same prompt.

What held it back

  • Camera moves stay conservative. On a "slow push-in along the forest path" prompt, the frame barely tightened across the clip; HappyHorse 1.1 and a couple of Seedance tiers on the same prompt pushed noticeably further into the scene. If you need a dramatic push, ask for large movement amplitude directly or expect to run it more than once.
  • No text-to-video or reference-to-video. This family only takes a starting image. If you need to write a shot from nothing, or hold a cast steady across reference photos, look elsewhere in our catalogue.
  • L4 Intense is the one soft spot. Two of three L4 Intense tries in our content-ladder run came back softened rather than as written; every other rung, including both L5 Extreme probe types, rendered every time.
  • It costs more than its siblings at matching resolution. At 540p, Q3 charges $0.07 a second against Q3 Turbo's $0.042 — 67% more for what looked, in our side-by-side runs, like comparable quality on most prompts.
  • Sound settings need a deliberate choice. Audio is on by default. Set generate_audio to false for a silent clip; bgm separately controls background music. The other Q3 tiers also expose audio generation.

What the results look like

Vidu Q3's page examples show its literal-animation range: a couple locked under a sheet with only the light moving across their backs, a woman caught in a gust of wind on a clifftop with her dress and hair streaming sideways, and a slow fall backward across a bed lit by blown-out morning light through a curtain — three different tests of how much a still can move while staying anchored to what the photo drew.

In our own head-to-head comparison, run the same day against HappyHorse 1.1, several Seedance and Wan tiers, and MiniMax H3, Q3's results were closest in spirit to the other Vidu tiers: careful, faithful to the source image, and unwilling to invent a bigger event than the prompt described. On a cat leaping onto a bookshelf, its camera stayed locked exactly where the still had it, and the landing kept the shelf's perspective intact — a detail that slipped in both Q3 Pro's and Q3 Turbo's take on the identical prompt, where the final frame tilted at an angle the opening frame never had. On a pour-over coffee shot, all three Q3 tiers produced nearly identical foam and steam detail, which tells us the quality difference between them is really about price and resolution ceiling, not raw output quality on straightforward physics.

Where it fell short of some rivals was camera scale: on the misty forest push-in, several other families closed the distance to the light source noticeably further than Q3 did in the same number of frames. If your shot depends on a big, obvious camera move, budget for a couple of takes or push movement_amplitude to large directly.

Prompts that worked

Vidu Q3 works well with a single clear instruction about what should move, aimed at a still image that already carries the wardrobe, setting and mood. Name the movement's mechanism (wind, light, a hand) before naming the effect it has.

1. A camera that just pushes in.

text
Slow push-in along the forest path as sunbeams break through the mist and a few leaves drift down.

Run this prompt on Vidu Q3 →

2. A subject turning toward camera, clothing described but not changed.

text
She turns from the window toward the camera, the silk dress catching the city light, and slowly runs her fingers through her hair.

Run this prompt on Vidu Q3 →

3. A physics-driven action with a mechanism named up front.

text
Hot water streams from the kettle spout into the dripper, steam rises and the coffee grounds bloom and bubble.

Run this prompt on Vidu Q3 →

Weak: "Make it move." With no mechanism or target named, the model has nothing to anchor a specific motion to. Strong: prompt 1 — one camera behavior, one visible cause (the mist, the sunbeams), nothing competing for attention.

How far it goes

Our content ladder runs from L1 Suggestive (lingerie, a slipping strap) through L2 Partial and L3 Strong to L4 Intense (an intimate scene between two adults) and L5 Extreme, tested as two separate probes: consensual restraint and impact play, and cinematic gore.

Across 19 runs between 14 and 25 September 2026, Vidu Q3 rendered every L1, L2, L3 and both L5 Extreme probe types as asked, with no refusals anywhere on the ladder. L4 Intense was the exception: one of three tries came back exactly as written, and two came back softened rather than refused outright. The page label is Uncensored: we add no content filter of our own here.

Not tested yet: two-women or two-men scenes above L1 Suggestive; the full 16-second duration ceiling on any content-ladder rung; text describing objects or props beyond what the starting image already shows.

Adults only, in every prompt and every starting image. A real, identifiable person needs to have consented, and you need the rights to any photo you animate.

Alternatives

How far Vidu Q3 goes

Each level, on each task, based on confirmed results. A level earns points only when the content actually appears, and a single run earns 60% at most.

Freedom points by level and task
LevelImage to video
MildThe lightest prompts in the test
3/3 delivered20.0 / 20 pts
ModerateWhere filtered models start to refuse
3/3 delivered20.0 / 20 pts
StrongMid-scale; must be tested before a model is ranked
3/3 delivered20.0 / 20 pts
Very strongWhere most models start to hold back
1/3 delivered13.3 / 20 pts
MaximumThe top of the scale, scored in two halves: class A and class B
Class A: 4/4 deliveredClass B: 3/3 delivered20.0 / 20 pts
Task score93.3 / 93.3Tested Sep 25, 2026

Tested Sep 14, 2026 to Sep 25, 2026

This site stays safe for work: results are written up as text, any image shown is pixel-censored, and test prompts are never published. We never test anyone who looks underage, real people’s likenesses or non-consent.

See every test record →See the uncensored leaderboard →

Run your own prompt and see how far it goes

Open the Playground

How Vidu Q3 earns its Spicy Index

Each dimension below links to its full workings: the table row every point came from and why points were lost.

Public leaderboards 45% of quality
Not yet tested

No exact match on a tracked public leaderboard yet. How it works

Blind output test 35% of quality
67.3
Arena quality 20% of quality
Not yet tested

Counts once the model has 100+ blind votes and a 95% interval under 200 Elo. How it works

Capability 10.5% weight
56.5
Freedom Score 30% weight
93.3

Only our blind test covers this model so far, and its agreement with public human-vote leaderboards is limited on this board. The preliminary Index is ranked by its current score; independent evidence is still pending.

Open the full test report →

Capability, item by item

Points from the public scoring table. Upscaled tiers and limits we set ourselves don’t count.

Native max resolution1080p
25.0 / 25
Longest single clip16 s
9.0 / 20
Native audioYes
15.0 / 15
Reference inputsNo reference images · no video or audio
0.0 / 15
Frame controlFirst frame only
5.0 / 10
Task coverageImage to video
2.5 / 10
Customization (LoRA)No LoRA support
0.0 / 5

Arena rating: 1031 (95% interval 892–1203); votes are still coming in.

What Vidu Q3 can do on SpicyAPI

Straight from the live model specification on SpicyAPI. Some creators offer more in their own apps; this is what you can call through our API today.

Capabilities by endpoint
FeatureImage to video
Resolutions540p · 720p · 1080p
Clip length1–16 s
Native audioYes
Seed controlYes
Negative promptNo
See every parameter and code sample →

How long it really takes

Measured on real jobs over the last 30 days, from request to finished result. Times include our own test runs.

Median time
82 s
90th percentile
3.3 min
Wait per second of video
16.2 s per output second
Success rate
Collecting data

A time appears once at least 10 completed jobs are in. Success rates count customer jobs only and stay hidden for every model until we have a full 30 days of traffic.

Models to consider instead of Vidu Q3

What do you think?

Was this review useful?

Log in to vote.

Comments

Comments are checked automatically and by our team. We remove spam, offensive content, personal attacks and anything that names a real person without consent. Report a comment and we review it within one business day.

Tried this model? Share what worked. You need a free account to comment, so we can keep spam out.

Vidu Q3 FAQ

Try Vidu Q3 yourself

Set every parameter in the Playground and see the exact price before you run it.

$0.75 per 5-second clip

More reviews

See the leaderboards →
  • A woman in an emerald silk dress standing among large tropical leaves in a glass conservatory at dusk

    Wan 3.0

    Pick Wan 3.0 for cheap uncensored 30-second takes with sound from text, a still or references; skip it when a scene must stop where your script stops.

    Spicy Index
    87.7
    Freedom Score
    96.0

    Tested Sep 24, 2026

    Read the review
  • A woman with wet hair and closed eyes lying back in a foam-filled stone bathtub by candlelight

    Seedance 2.5 Spicy

    Pick Seedance 2.5 Spicy for the harder rungs or 4K from a still: it went further than base 2.5 in our tests, but base 2.5 costs less.

    Spicy Index
    67.2
    Freedom Score
    96.7

    Tested Sep 25, 2026

    Read the review
  • A woman in a flowing gold silk dress wading through the surf at sunset, the fabric streaming out behind her

    Seedance 2.5

    Pick Seedance 2.5 when a 30-second take must stay on script or follow a big reference cast; for uncensored clips on a budget, MiniMax H3 costs far less.

    Spicy Index
    81.4
    Freedom Score
    83.6

    Tested Sep 24, 2026

    Read the review