Gemini Omni 1.1 Flash AI Video Generator
Create 4 to 10 second videos with Google Gemini Omni 1.1 Flash on HeyMarmot. Direct the shot with up to 7 reference images or a reference clip, lock the opening and closing frame, and export at up to 4K.
Optional: add up to 7 reference images, or one reference clip alongside images. First and last frame mode uses 1 to 2 images, and audio references are not supported yet.
Up to 4K
Render the same direction at draft speed or full 4K detail. Test the framing at 360P, then re-run the shot at 1080P or 4K once the direction is right.
Ways to direct the shot
One directed shot
16:9 or 9:16, 4 to 10 seconds
Gemini Omni 1.1 Flash and the best AI video models in one studio
Start with Gemini Omni 1.1 Flash for multimodal direction and 4K output, then compare the same idea against other leading video models without switching tools.
Show the model what you mean, not just tell it
Gemini Omni 1.1 Flash reads up to seven reference images or a reference clip alongside your prompt, so character, wardrobe, set and lighting stay consistent from the first frame to the last.
Create a Gemini Omni 1.1 Flash video in three steps
Move from a creative brief to a finished clip without leaving the page.
Describe and reference
Write the scene, then add the reference images or clip that should guide the generation.
Choose the model and settings
Select Gemini Omni 1.1 Flash, then set duration, aspect ratio and resolution.
Generate and refine
Review the result, download it, or adjust the direction and create another version in the same conversation.
Choose the right model for every shot
Use Gemini Omni 1.1 Flash, Wan 3.0, Hailuo H3, Seedance 2.5, Veo 3.1 and more from one workspace and one credit balance.
Gemini Omni 1.1 Flash
Available in the HeyMarmot video workspace.
Wan 3.0
Available in the HeyMarmot video workspace.
Hailuo H3
Available in the HeyMarmot video workspace.
Seedance 2.5
Available in the HeyMarmot video workspace.
Veo 3.1
Available in the HeyMarmot video workspace.
Sora 2
Available in the HeyMarmot video workspace.
Kling 3.0
Available in the HeyMarmot video workspace.
What creators make with Gemini Omni 1.1 Flash
Frame control, multimodal references and 4K output make the model useful from concept tests through to delivery.
Cinematic shots
Set the opening and closing frame, then let the model fill in the camera move and the performance in between.
Social video
Generate 9:16 hooks, product moments and short explainers at the pace vertical feeds expect.
Product and brand
Feed in packshots and brand references so material, colour and logo placement survive the generation.
Character consistency
Reuse the same reference set across shots to keep a character recognisable throughout a longer edit.
A complete Gemini Omni 1.1 Flash workflow
HeyMarmot keeps model settings, references, results and iteration history in one thread so you can move from test to final render quickly.
Multimodal prompting
Combine a prompt of up to 20,000 characters with as many as 7 reference images, or one reference clip.
First and last frame
Supply an opening image, and optionally a closing image, when the composition at each end has to be exact.
360P to 4K
Draft at low resolution to test direction, then re-render the same shot at 1080P or 4K.
4 to 10 seconds
Pick the clip length that fits the beat, in 4, 6, 8 or 10 second steps.
Reusable asset library
Keep references and generated clips ready for the next variation or a new conversation.
Conversation history
Prompts, settings, source files and every Gemini Omni result stay connected in one thread.
Gemini Omni 1.1 Flash FAQ
Practical answers about creating Gemini Omni 1.1 Flash video on HeyMarmot.
What is Gemini Omni 1.1 Flash?
Gemini Omni 1.1 Flash is a Google multimodal video generation model. It creates video from text alone, from an opening and closing frame, or from image and video references, and it is available in the HeyMarmot video workspace.
What inputs can I use with Gemini Omni 1.1 Flash?
You can generate from a prompt alone, add up to 7 reference images, or add one reference clip together with images. First and last frame mode takes 1 or 2 images. Audio references are not supported yet.
How long can a Gemini Omni 1.1 Flash video be?
Each generation is 4, 6, 8 or 10 seconds. Longer sequences are built by generating several shots and editing them together.
Which resolutions and aspect ratios are supported?
Gemini Omni 1.1 Flash supports 360P, 720P, 1080P and 4K output, in 16:9 or 9:16.
How does first and last frame mode work?
Upload one image to fix the opening frame, or two images to fix both ends. In this mode the model does not accept additional image or video references.
How many credits does a generation cost?
Credits depend on resolution, duration and whether you add a reference video. A 4 second clip without a reference video starts at 450 credits, and 4K generations cost more. The exact cost is shown before you submit.
Keep exploring AI video
Browse examples, compare workflows or check credit pricing before your next generation.