Vidu Q4
Flagship multi-ref
i2v / r2v · 3–16s · up to 4K · native audio
Video · Vidu
Start with Q4 for flagship multi-ref and 4K, or pick Turbo / Pro / Mix for Q3 workflows — then create on-site with the model locked.
Vidu on sora2.cc covers Shengshu video models synced from the catalog: Q4 preview plus the Q3 ladder (Turbo, Pro, Pro Fast, base, Mix).
Q4 emphasizes image-to-video and up to 15 references with native audio and 540p–4K. Q3 Turbo covers four input modes; Pro focuses text and frames; Mix leans denser multi-ref.
Compare here, open a tier page for detail, then deep-link into /generate/.
Climb for flagship multi-ref; pick Turbo for full modes; Mix when refs matter most.
Flagship multi-ref
i2v / r2v · 3–16s · up to 4K · native audio
Full modes
t2v / i2v / start-end / r2v with audio
Quality frames
t2v / i2v / start-end up to 1080p
Dense multi-ref
Higher-credit reference packs at 720p/1080p
Vidu Q4 NEW
Vidu Q4 — i2v / r2v, 3–16s, 540p–4K, native audio
Open model page →
Vidu Q3 Turbo
Vidu Q3 Turbo — t2v / i2v / start-end / r2v, audio
Open model page →
Vidu Q3 Pro
Vidu Q3 Pro — t2v / i2v / start-end, up to 1080p
Open model page →
Vidu Q3 Pro Fast
Vidu Q3 Pro Fast — image-to-video, 720p / 1080p
Open model page →
Vidu Q3
Vidu Q3 — reference-to-video, 1–7 refs, audio
Open model page →
Vidu Q3 Mix
Vidu Q3 Mix — multi-ref video, 720p / 1080p
Open model page →
Ship short clips with optional native audio for feeds and openers.
Animate product or character stills with i2v or first-frame modes.
Lock identity across multi-ref packs on Q4, Q3, or Mix.
Use start-end on Turbo / Pro when both ends of the shot matter.
Need 4K multi-ref → Q4; need t2v → Turbo/Pro.
Read capabilities, then use Create to deep-link /generate/.
Confirm credits, run, then refine or switch siblings.
Common Vidu intents:
Modes, durations, and audio defaults follow the live catalog per tier and can change with runtime updates.