Gemini Omni

Gemini Omni is described as Google DeepMind's multimodal video creation model, announced at Google I/O 2025. The page presents it as supporting multi-turn conversational editing, combined text, image, video and audio input, and real-world context. These are product descriptions, not a guarantee of output accuracy or a statement that every capability is available in the selected model. Sign in to Nano Banana Pro, choose a supported model and check its credit cost before generating.

Sign in to Nano Banana Pro, select an available video model and check your account's credit balance and the displayed generation cost. Enter a text prompt or add supported references, then submit your request. Available models, settings and credit costs may vary; check them before generating.

The page describes text prompts, up to 7 reference images, one video clip of up to 100 MB and 30 seconds, and audio inputs. It also describes combining input types, such as pairing an image reference with text instructions. These limits and inputs may depend on the selected model and configured mode, so check the options shown before generating.

Start with an initial generation, then use natural-language instructions to request changes across multiple turns, such as adjusting the camera angle or lighting, replacing objects, adding effects or changing the style. The page describes edits as building on the previous result and aiming to retain elements you do not mention. Results can vary with the input and instructions.

The page describes 4, 6, 8 or 10-second videos in 16:9 landscape and 9:16 portrait formats, along with seed control. These are not guaranteed settings for every model. Check the selected model's available options before generating.

Commercial-use rights depend on the applicable service terms, selected model and your account. Review those terms before using generated videos commercially; this page does not establish a blanket commercial-use licence.

This page presents Gemini Omni's potential distinctions as multi-turn conversational editing, combined text, image, video and audio inputs, and use of real-world context. Capabilities and available settings vary by model, and the descriptions here do not verify comparative performance or guarantee physically accurate or contextually correct results. Check each model's options and terms for your intended task.

Generation mode: Text-to-Video / Image-to-Video / Reference-to-Video / Video-to-Video. Suitable for high-quality video projects, marketing content, social videos, product demonstrations and creative storytelling with image references.

Prompt tips: Describe the lighting, camera movement and atmosphere in detail. Use reference images to guide visual consistency. Choose a duration from the options available for the selected model. Start with one subject, one action and a clear camera direction. Compare the result with your brief; generated motion can vary.

Duration: 4s / 6s / 8s / 10s. Check the controls for the active generation mode before submitting.

Resolution: 720p / 1080p / 4k. Check the controls for the active generation mode before submitting. Output quality can affect the credit estimate. Confirm the displayed total before generating.

Aspect ratio: 16:9 / 9:16. Check the controls for the active generation mode before submitting. Choose the framing for your destination and check how uploaded images are cropped.

Generation mode: Text-to-Video / Image-to-Video / Reference-to-Video / Video-to-Video. Reference image: 7. Video references are supported. Only use source material you can process, and say what each reference should guide. A reference cannot guarantee identical details. Watch the entire clip for continuity, unintended objects, timing and audio where available before using it.