videoby Google DeepMindAvailable in Cinemagiq
Gemini Omni 1.1 Flash
Video with dialogue and ambience, from 360p drafts to 4K
A 10-second clip costs from about $0.60
Video with dialogue and ambience, from 360p drafts to 4K
A 10-second clip costs from about $0.60
Gemini Omni 1.1 Flash is a video model from Google DeepMind. In Cinemagiq it generates picture and sound together, including dialogue and ambience, from a prompt, a first frame, a first and last frame, or reference images. It renders from quick 360p drafts all the way up to 4K, so one model can carry a shot from first test to final delivery.
Write a prompt for text-to-video, or add a first frame, or a first and last frame, to pin the shot. Up to 6 reference images can join those frames in the same job. Clips run 3 to 10 seconds in 16:9 or 9:16, at 360p, 720p, 1080p or 4K. Sound is always on, so the prompt is where you set the spoken lines, effects and room tone.
Pick Gemini Omni 1.1 Flash for short dialogue moments and atmosphere-driven shots, such as a line spoken by a campfire or a quiet cafe scene. Draft at 360p to check timing and performance, then render the keeper at 1080p or 4K. For shots longer than 10 seconds, use Seedance 2.5 or Kling O3. For wide 21:9 frames, pick a model that offers that ratio.
Made in Cinemagiq with Gemini Omni 1.1 Flash. The exact prompt is under each one.
PromptMedium close-up of an old man sitting by a campfire at night. He says, 'Every mountain here has a story.' The fire crackles and sparks rise into the dark sky.
PromptA barista pours steamed milk into a latte and draws a heart, the camera slowly tilts down. Sound: quiet morning cafe ambience and the hiss of the steam wand.
Straight from the model list Cinemagiq runs today.
| Variant | What drives the shot | Length | Resolution | Sound |
|---|---|---|---|---|
| Gemini Omni 1.1 FlashText, frames or references drive the shot, 360p to 4K with audio | Prompt, first frame (optional), last frame | 3 to 10 s | 360p, 720p, 1080p, 4K | Always on |
Gemini Omni 1.1 Flash is a video model from Google DeepMind. In Cinemagiq it makes 3 to 10 second clips with sound from a prompt, frames or up to 6 reference images, at 360p up to 4K, in 16:9 or 9:16.
Yes. Sound is always on and can include dialogue and ambience. Write the spoken line in quotes, name who says it, and describe the background sound. Keep lines short so they fit comfortably in a clip of 10 seconds or less.
Clips run from 3 to 10 seconds per generation. For a longer moment, split it into several shots on the Production Board, or pick Seedance 2.5, which reaches 30 seconds in one generation.
Yes. Up to 6 reference images can work alongside a first frame, or a first and last frame, in the same job. Use the frames to fix the composition and the references to keep characters, costumes or props consistent.
Generate with Gemini Omni 1.1 Flash and the other models Cinemagiq runs in one AI-native workspace, and pick the right one per shot.