Hosted video models, hands on

Member lesson

Gemini Omni Flash, and editing a clip by talking to it

Lesson 5 of 10 in Hosted video models, hands on.

Checked against the makers' documentation on 25 September 2026.

Introducing Gemini Omni: Create Anything from Anything · Google
Introducing the Gemini Omni Flash API · Sam Witteveen
Gemini Omni Flash 1.1 Is Here — Here’s What’s NEW · ElevenLabs
Lesson

Gemini Omni Flash is Google's other video model, and the one you can talk to. You give it text, pictures or a short video, it returns a clip of 3 to 10 seconds with sound at 24 frames a second, and then you can send a sentence such as 'Change the light to late-afternoon sun. Keep everything else the same.' and get back the same clip with that one thing changed. It is called through the Interactions API as gemini-omni-1.1-flash, with gemini-omni-flash-preview beside it, and our key reaches both. Veo cannot do this. A Veo clip is finished when it arrives, and changing it means generating it again.

turn 1 · prompt → clip aturn 2 · warmer light, keep the restturn 3 · no steam, keep the restextend · +3 to 10 s, 40 s in all360p draft720p own render1080p / 4k upscaledeach turn is a new render and a new bill
Talking to a clip with Gemini Omni: a first draft at 360p, then turns that change one thing and keep the rest, then an extension of three to ten seconds up to forty; 720p is Omni's own render, 1080p and 4K are upscaled.
An ink drawing of a potter's wheel with a green bowl on it and three earlier versions of the same bowl in a row on a shelf, each slightly changed.
One bowl reshaped in small named steps, instead of thrown again from scratch.
Member

The rest of this is for members.

You have read the opening of this lesson. The rest of it comes with a membership.

One membership covers all four courses and Brands.

  • All lessons in the four Learn courses
  • All brand breakdowns in the Brands section
  • All quizzes, flashcards and crosswords
  • An AI teacher on every lesson, for up to 60 questions a day
  • All course material published during the membership

Price

Membership is from $12 a month on the Yearly plan, which is charged at $144 once a year. Lifetime access is $59, paid once.Membership is from ₹200 a month on the Yearly plan, which is charged at ₹2,400 + GST once a year. Lifetime access is ₹3,999 + GST, paid once.

Check what stuck

Members only

You can't take this check yet.

Members answer these 3 questions and keep the ticket for this lesson. Every lesson has one.

  1. Question 1, locked

    Which of these words means this? “Google's endpoint for multi-turn work. Omni Flash is called here, not through generate_videos.”

    • Retry
    • Interactions API
    • Shot type
    • Dialogue
  2. Question 2, locked

    Which of these words means this? “The id of your last turn. Send it back and Omni edits that clip instead of starting again.”

    • Polling
    • Presigned link
    • previous_interaction_id
    • Sound cue
  3. Question 3, locked

    Which of these words means this? “Enlarged after rendering. Google's own table marks Omni's 1080p and 4K this way.”

    • Retry
    • Shot type
    • Dialogue
    • Upscaled

Elsewhere in the library

Hear about new courses.

Leave an email and we will tell you when something new lands. We do not write often.