VideoGenReview

What each AI video model's release notes say changed

Vidu changes by date, 2025 to 2026

Vidu: 7 entries, 28 July 2025 to 14 September 2026, naming 2 versions from Vidu Q2 to Vidu Q3; 4 name no model. Apart from new models, most concern references or consistency and sound, voice or lip-sync. As of 2026-09-25.

Entries per line, this one markedA long bar means a detailed log, not necessarily a faster-moving model.Entries per line, this one markedKling AI3737Runway3131Wan2626PixVerse2020LTX1616Seedance1616Google Veo1515Luma Ray99Adobe Firefly Video77MiniMax Hailuo77Vidu77HappyHorse66Amazon Nova Reel55Gemini Omni55Grok Imagine44Midjourney video44Sora44
Fig. 1 A long bar means a detailed log, not necessarily a faster-moving model.

Inclusion rule. All seventeen lines on the register, each with its count of dated entries and version rows. Order. Alphabetical by line.

1What the notes say

Vidu's API update notice dates six video changes between July and November 2025: lip-sync, an off-peak mode, reference video from one to seven images, the Q2 Pro and Q2 Turbo models, extension and text-to-video for Q2, and sound and picture in one output. Vidu Q3 reference models for drama and for ads appear on Alibaba Cloud Model Studio's list in September 2026.

2Every entry

Vidu versions by first dated entryOne dot per named version, placed at its first dated entry.Vidu versions by first dated entry2025-09-24Vidu Q22026-09-14Vidu Q3
Fig. 2 One dot per named version, placed at its first dated entry.
Every Vidu entry, oldest first.
DateVersionWhat changedSource
28 July 2025Not named in the noteLip-sync from text or audioMaker's note
7 August 2025Not named in the noteAn off-peak mode at lower credit costMaker's note
26 August 2025Not named in the noteReference video generation from one to seven imagesMaker's note
24 September 2025Vidu Q2Vidu Q2 Pro and Q2 Turbo for image-to-video and start-and-end framesMaker's note
21 October 2025Vidu Q2Q2 gains reference-to-video, text-to-video and extensionMaker's note
13 November 2025Not named in the noteAudio and video in one output for reference-to-video and image-to-videoMaker's note
14 September 2026Vidu Q3Vidu Q3 reference-to-video models for drama and for ads are listed on Model StudioArrival on Alibaba Cloud Model Studio

Inclusion rule. All entries on the register for this line: the maker's notes, platform arrivals and undated rows. Order. By date, oldest first; undated rows last.

3Each entry with its source

  • July 28
    Lip-sync from text or audio2025Vidu, API update notice / since 2025-07-28 / checked 2026-09-25
  • August 7
    An off-peak mode at lower credit cost2025Vidu, API update notice / since 2025-08-07 / checked 2026-09-25
  • August 26
    Reference video generation from one to seven images2025Vidu, API update notice / since 2025-08-26 / checked 2026-09-25
  • September 24
    Vidu Q2 Pro and Q2 Turbo for image-to-video and start-and-end frames2025Vidu, API update notice / since 2025-09-24 / checked 2026-09-25
  • October 21
    Q2 gains reference-to-video, text-to-video and extension2025Vidu, API update notice / since 2025-10-21 / checked 2026-09-25
  • November 13
    Audio and video in one output for reference-to-video and image-to-video2025Vidu, API update notice / since 2025-11-13 / checked 2026-09-25
  • vidu/viduq3-drama_reference2video
    Vidu Q3 reference-to-video models for drama and for ads are listed on Model StudioAlibaba Cloud Model Studio, newly released models / since 2026-09-14 / checked 2026-09-25

4Versions

  • Vidu Q2: 2 entries, from 24 September 2025
  • Vidu Q3: 1 entry, from 14 September 2026
  • Not named in the note: 4 entries

5Kinds of change

6Latest dated entry

14 September 2026, Vidu Q3: Vidu Q3 reference-to-video models for drama and for ads are listed on Model Studio.

A Vidu reference-to-video model designed for drama and AI comic series production.

Alibaba Cloud Model Studio, newly released models

7Neighbouring lines

8Notes read

How entries are chosen and sorted: the method.