Gemini Omni 1.1 Flash turns production prompts into short landscape or portrait clips at a selected output resolution.
Use text alone, or add optional image and video references to guide subjects, composition, visual language, or motion.
Final usage is the selected resolution rate multiplied by output duration.
Similar Models
Create videos up to 30 seconds, combine as many as 50 multimodal references, and refine specific moments with frame-level control.
A multimodal video generation model for text prompts, first-frame animation, and reference-guided production clips.
Create 4 to 10-second videos from text, up to seven reference images, or one short reference video with synchronized sound.
Model overview
Gemini Omni 1.1 Flash on sjolt provides text-to-video and reference-to-video workflows for 4, 6, 8, 10-second clips. Choose 16:9 or 9:16 framing, select one of four output resolutions, and use the same inputs through the asynchronous task API.
Select 360p, 720p, 1080p, or 4K for each generation. API requests use 360p, 720p, 1080p, and 4k, while the interface presents the highest tier as 4K.
Add up to 7 PNG, JPG, JPEG, or WebP images and optionally one MP4 up to 10 seconds. References can guide subject appearance, composition, visual treatment, or motion while the prompt defines the intended scene.
Submit a text or reference request for 4, 6, 8, 10 seconds, keep the returned task ID, and query task status until the generated video URL is available. An optional callback URL can receive the terminal task result.
Model characteristics
Direct the scene, action, camera, visual treatment, and sound cues from one required prompt.
Guide the result with up to 7 images and one MP4 no longer than 10 seconds.
Choose 16:9 or 9:16 framing, 4, 6, 8, 10 seconds, and one of four resolutions.
Submit a task, poll its status, and retrieve generated video URLs from the successful result.
FAQ
Gemini Omni 1.1 Flash is a Google video model available through separate text-to-video and reference-to-video task routes on sjolt.
Use text-to-video for prompt-only generation or reference-to-video when optional images or a short MP4 should guide the result.
No. The reference route accepts a prompt without media, matching the optional-reference workflow in the Playground.
Use up to 7 PNG, JPG, JPEG, or WebP images and one MP4 no longer than 10 seconds.
The available output resolutions are 360p, 720p, 1080p, and 4K. API requests use `4k` for the 4K option.
Choose 4, 6, 8, 10 seconds and 16:9 or 9:16 framing.
Submit a request with a prompt and supported settings, poll the returned task ID, then read generated video URLs from a successful result.
