gen‑ai.news
ClosedFreemium

Vidu

Shengshu Technology

Vidu is a video generation model line from Shengshu Technology. The latest release is Vidu Q4 (7 Oct 2026). This page tracks all 7 Vidu releases since 30 Jul 2024, with specs, licensing, and how to access each one. Official site →

Vidu Q4released

ClosedFreemiumOfficial APIConsumer app

Adds native audio generation alongside improvements to motion quality and prompt adherence

License
Proprietary
Availability
Official API, Consumer app
Audio
native audio generation (synchronized sound produced in single generation pass)
Motion Quality
improved
Prompt Adherence
improved

First Vidu release to integrate native audio directly into the video generation pipeline rather than as a post-processing add-on.

Vidu S1released

ClosedFreemiumOfficial APIConsumer app

Real-time interactive video foundation model: streams video continuously and turns a single uploaded image into a character driven live by voice input - a different product category from the Q-series clip generators

License
Proprietary
Commercial use
Allowed
Availability
Official API, Consumer app
Frame rate
25 fps continuous, up to 42 fps
Resolution
540p

Unveiled at the 2026 Global Digital Economy Conference, Beijing.

Vidu Q3released

ClosedFreemiumOfficial APIConsumer app

Shengshu (with Tsinghua's TSAIL lab) claims the first video model to generate native audio and video jointly in one pass, up to 16s at 1080p. Reference-to-Video with intelligent camera switching launched globally 2026-04-13

License
Proprietary
Commercial use
Allowed
Availability
Official API, Consumer app
Audio
native audio and video generated together in one pass
Resolution
1080p
Max duration
16s

Base model 2026-01-30; announcement_url is the 2026-04-13 Q3 Reference-to-Video press release, the only official English-language page located.

Vidu Q2released

ClosedFreemiumOfficial APIConsumer app

Nuanced emotional expression, smarter cinematography, faster generation; Flash mode (~20s) and Cinematic mode. Reference-to-Video with up to 7 reference images followed 2025-10-21; Q2 image generation with unlimited free 1080p later in 2025

License
Proprietary
Commercial use
Allowed
Availability
Official API, Consumer app
Max duration
8s

Vidu Q1released

ClosedFreemiumOfficial APIConsumer app

1080p with cinematic VFX transitions and auto BGM/SFX; multi-reference (up to 7 images) added July 2025

License
Proprietary
Commercial use
Allowed
Availability
Official API, Consumer app
Resolution
1080p

Announced 2025-03-29; global launch 2025-04-21.

Vidu 2.0released

ClosedFreemiumOfficial APIConsumer app

Sub-10s generation at very low cost; from Shengshu Technology (Tsinghua)

License
Proprietary
Commercial use
Allowed
Availability
Official API, Consumer app

Vidu 1.0released

ClosedFreemiumOfficial APIConsumer app

China's first Sora-class model; up to 16s

License
Proprietary
Commercial use
Allowed
Availability
Official API, Consumer app
Resolution
1080p
Architecture
U-ViT
Max duration
16s

Unveiled 2024-04-27; public/global launch 2024-07-30.

FAQ

When was Vidu Q4 released?
Vidu Q4 was released on 7 Oct 2026.
Is Vidu open source?
No. Vidu Q4 is closed: the weights are not released and it is available through an API or app only.
Where can I use Vidu?
Vidu Q4 is available via Official API and Consumer app.

See Vidu in context on the video model timeline, or browse all model families.