Skip to content

Changelog: Gemini Omni Flash. One prompt in, a whole scene out

2026-06-30 · changelog: model release · Higgsfield · original

Topic: Video generation & shot direction (also: Audio, voice & music) · mode: Release note · tags: vendor claims

Higgsfield's changelog of 30 June 2026 brings Google DeepMind's Gemini Omni Flash to Higgsfield in Preview (beta): it composes text, images and video into one shot, keeps characters across edits, and changes one element of a clip by a plain-language instruction; 720p clips of 4 to 10 seconds with native audio.

What is said

  • Gemini Omni Flash: one prompt in, a whole scene out

What is said

★ — how much it matters to the source.

  • ★★★ Gemini Omni Flash from Google DeepMind reads text, images and video together and composes them into one coherent shot, so a scene is built from references rather than described from scratch. · Gemini Omni Flash
  • ★★ Higgsfield says Gemini Omni Flash holds a character's face, outfit and voice across every edit, so a shot can be revised without recasting it. · Gemini Omni Flash
  • ★★ Higgsfield says gravity, weight and collisions behave correctly in Gemini Omni Flash, so motion looks filmed rather than simulated. · Gemini Omni Flash
  • ★★★ Gemini Omni Flash changes one element of an existing clip from a plain-language instruction instead of regenerating the whole scene. · Gemini Omni Flash
  • ★★ To start with Gemini Omni Flash on Higgsfield: Video → model selector → Gemini Omni, then add a prompt, or bring an image, a start and end frame, or a clip to edit. · Gemini Omni Flash, keyframe
  • ★★★ Gemini Omni Flash arrives on Higgsfield in Preview (beta): 720p clips of 4, 6, 8 or 10 seconds, in 16:9 or 9:16, with native audio. · Gemini Omni Flash, native audio

Threads in this video

  • launched · Gemini Omni Flash on Higgsfield — Gemini Omni Flash arrives on Higgsfield in Preview (beta): 720p clips of 4, 6, 8 or 10 seconds, in 16:9 or 9:16, with native audio.

Where the stories that moved here went next

Mentioned

Central: Gemini Omni Flash

Also: Google DeepMind · keyframe · native audio

← Previous: 5 Best Ways to Access Gemini Omni Flash in 2026: Platforms and Plans · Next → How to Use Gemini Omni Flash for Multi-Shot AI Video Production