Video · ShengShu

Vidu Q3

ShengShu's January 2026 model for short stories: 16 s, several camera angles and sound in one pass. Pick it for a small scene with dialogue.

Anonymized Moderate cost
Latest versionVidu Q3January 30, 2026
Artificial Analysis Image to Video Arena1,056 Elo#11 of 22 on Artificial Analysis Image to Video Arena
Price$0.0872 per second of video$0.436 a 5 s clip
Key specClips up to 16 s, with sound

What sets it apart

Vidu Q3 from ShengShu generates up to 16 s in one pass at 1080p with dialogue, effects and music, and detects scene boundaries to cut between camera angles itself. A reference-to-video mode arrived in April 2026.

  • 16 s in one pass, 1080p
  • Automatic scene cuts
  • Pro and Turbo variants

Strengths and limits

Where it shines

  • Switches camera angles inside one generation
  • Dialogue, effects and music generated with the picture

Think twice when

  • Mid-range to high price per second
  • No reference mode here
  • 3 to 16 s, native sound
  • 360p to 1080p

Use cases

Small scene with dialogue

Two neighbors talk over a garden fence. He says 'Lovely tomatoes this year.' She answers 'Secret is the compost.' Cuts between both faces. 12 seconds.

It cuts between angles on its own and syncs lips.

Start and end frame animation

From my first frame (closed door) to my last frame (door open, warm light), slow push-in with creaking sound.

Image mode accepts configurable start and end frames.

Camera move on a landscape

A tracking shot along a coastal road at sunset, convertible driving, ocean wind sound, 10 seconds.

Native camera control handles push-ins and tracking.

Against its rivals

  • PixVersePixVerse C1 targets action shots at a lower price.#20 of 22 on Artificial Analysis Image to Video Arena (963 Elo)
  • Kling V3Kling V3 runs 15 s with 4K and motion control.#12 of 22 on Artificial Analysis Image to Video Arena (1,055 Elo)
  • Seedance 2.0 and 1.5Seedance 2.0 accepts references and audio inputs.#18 of 22 on Artificial Analysis Image to Video Arena (1,000 Elo)
Compare side by side ↗

Version history

Each version, dated, with what it changed over the one before.

  1. Vidu Q3Latest
    • Up to 16 s of continuous video at 1080p and 24 fps in a single pass.
    • Native audio (dialogue, effects, music) and automatic multi-shot scene detection.
    • Reference-to-video mode added in April 2026.
    Source : developer.puter.com

Benchmarks

How to read these scores: what a benchmark tells you.

Artificial Analysis Image to Video Arena: #11 of 22 models measured in the catalog. The same blind duel, starting from one image: which clip brings it to life better. An Elo rating: higher is preferred more often. 100 points apart means the higher-rated model wins about 64% of duels.

Artificial Analysis Image to Video ArenaElo rating · Artificial Analysis, October 4, 2026
  1. 1. MiniMax H3 MaxMiniMax H3 (Hailuo)1,195
  2. 2. MiniMax H3MiniMax H3 (Hailuo)1,181
  3. 3. Wan 3.0Wan 3.01,164
  4. 4. HappyHorse 1.1HappyHorse1,104
  5. 5. Grok Imagine 1.5Grok Imagine video1,098
  6. 6. HappyHorse 1.0HappyHorse1,083
  7. 7. Veo 3.1 Full QualityVeo 3 and 3.11,082
  8. 8. Wan 2.7Wan 2.x1,077
  9. 9. Grok ImagineGrok Imagine video1,072
  10. 10. Veo 3.1 FastVeo 3 and 3.11,066
  11. 11. Vidu Q3Vidu Q31,056
See the numbers
ModelElo
1. MiniMax H3 Max1,195
2. MiniMax H31,181
3. Wan 3.01,164
4. HappyHorse 1.11,104
5. Grok Imagine 1.51,098
6. HappyHorse 1.01,083
7. Veo 3.1 Full Quality1,082
8. Wan 2.71,077
9. Grok Imagine1,072
10. Veo 3.1 Fast1,066
11. Vidu Q31,056

All 2 models in this family

ModelUsePrivacyPrice
Vidu Q3vidu-q3-text-to-videoText to videoAnonymized$0.0872 per second of video$0.436 a 5 s clip
Vidu Q3vidu-q3-image-to-videoImage to videoAnonymized$0.0872 per second of video$0.436 a 5 s clip

Specifications

SpecificationVidu Q3
ReleasedJanuary 30, 2026
WeightsClosed
Acceptstext, images
Producesvideo, audio
Clip length3 to 16 s, with sound
Resolutions360p, 540p, 720p, 1080p
Price$0.0872 per second of video
PrivacyAnonymized

Prices, context and lengths from the live catalog on October 16, 2026; parameters and licenses from the maker.