VideoGenReview

What each AI video model's release notes say changed

When names in the notes do not match

The same model can appear under several names. Google's changelog calls it gemini-omni-1.1-flash; Higgsfield calls it Gemini Omni 1.1 Flash. Runway's developer API added a model it names MiniMax Hailuo 3.0 five days after MiniMax announced H3, and MiniMax's own notes do not use the name Hailuo 3.0. From 4 source pages, read 2026-09-25; no maker pays to be listed.

Where this note's examples come fromThe examples in this note come from 5 entries across 2 model lines.ExamplesGemini Omni2 entriesMiniMax Hailuo3 entries
Fig. 1 Each example links to its line page.

The register files an entry under the name its note uses and does not merge names the makers have not merged. Where Higgsfield calls a model a variant, as with MiniMax H3 Max, it is kept as its own version.

The entries this note draws on, read 2026-09-25.
DateLineVersionWhat changed
31 July 2026MiniMax HailuoMiniMax H3MiniMax H3 reads text, image, video and audio together as creative context
5 August 2026MiniMax HailuoHailuo 3.0Runway's developer API adds a model it names MiniMax Hailuo 3.0
27 August 2026Gemini OmniGemini Omni 1.1 FlashGemini Omni 1.1 Flash goes GA with extension, first-and-last-frame input and 360p to 4K output
28 August 2026Gemini OmniGemini Omni 1.1 FlashGemini Omni 1.1 Flash on Higgsfield, up to 10 seconds and 4K
31 August 2026MiniMax HailuoMiniMax H3 MaxMiniMax H3 Max, a variant tuned for prompt adherence, on Higgsfield

Inclusion rule. Entries used as examples in this note. Order. By date, oldest first; undated rows last.

1In the notes

It understands creative intent across multimodal context — text, image, video, and audio — and delivers more natural, coherent generation and expression.

MiniMax, model release notes

MiniMax Hailuo 3.0 is now live on Runway Dev.

Runway, changelog

Resolution control : New resolution parameter in video_config supports 360p , 720p (default), 1080p , and 4k outputs.

Google, Gemini API changelog

Up to 10 seconds per generation , at 360p, 720p, 1080p, or 4K.

Higgsfield, changelog

a post-trained variant of MiniMax H3 tuned for prompt adherence and aesthetics.

Higgsfield, changelog
  • MiniMax H3
    MiniMax H3 reads text, image, video and audio together as creative contextMiniMax, model release notes / since 2026-07-31 / checked 2026-09-25
  • MiniMax Hailuo 3.0
    Runway's developer API adds a model it names MiniMax Hailuo 3.0Runway, changelog / since 2026-08-05 / checked 2026-09-25
  • August 27
    Gemini Omni 1.1 Flash goes GA with extension, first-and-last-frame input and 360p to 4K output2026Google, Gemini API changelog / since 2026-08-27 / checked 2026-09-25
  • Gemini Omni 1.1 Flash
    Gemini Omni 1.1 Flash on Higgsfield, up to 10 seconds and 4KHiggsfield, changelog / since 2026-08-28 / checked 2026-09-25
  • MiniMax H3 Max
    MiniMax H3 Max, a variant tuned for prompt adherence, on HiggsfieldHiggsfield, changelog / since 2026-08-31 / checked 2026-09-25

2Where new versions end up

Release notes such as “When names in the notes do not match” matter once a pipeline runs the models; SceneMixer's guides describe how it does:

AI dubbing replaces the audio of a finished clip and then tries to match the lips. Native-language generation gives the video model the line in the target language before the shot exists, so mouth movement, pauses and shot length are produced for that language.

SceneMixer, Native-Language Dialogue vs. Dubbing in AI Video (2026): Why the Cast Should Speak the Line — “Is native-language generation the same as AI dubbing?” (read 2026-10-13)

SceneMixer writes dialogue into the script analysis, assigns each line to a shot with a speaker and a short delivery note, and the video model performs it — picture and audio generated together, so the mouth forms that sentence in the language you chose.

SceneMixer, AI Dialogue Generator (read 2026-10-13)

3Other reading notes

4Notes read