Home 零对冲(ZeroHedge)每日HackerNews

Gemini Omni

原始链接: https://deepmind.google/models/gemini-omni/

相关文章
原文
Google DeepMind DeepMind
Build with Gemini
Try Gemini

Gemini Omni

Create anything from any input – starting with video

Try in Gemini
Try in Google Flow
Learn how to prompt

Edit any video through natural, step-by-step conversation

Think of Gemini Omni like Nano Banana, but for video. Every edit you make builds on the one before – maintaining a consistent, coherent scene.

Try in Gemini
Try in Google Flow

Apply real world knowledge

Gemini Omni combines an intuitive understanding of physics with Gemini's knowledge of history, science, and cultural context – bridging the gap from photorealism to meaningful storytelling.

Try in Gemini
Try in Google Flow

Reference anything

Turn any reference—image, text, video, or audio—into a single, cohesive output.

Try in Gemini
Try in Google Flow
Explore the latest

Gemini Omni is where Gemini’s ability to reason meets the ability to create. It delivers a leap in world understanding, multimodality, and editing.

View prompt guide

Edit through natural conversation

Think of Gemini Omni like Nano Banana – but for video. Build and fine-tune your creation at any step with natural language.

Transform your world

Change the aesthetic, action, or effect based on your input video.

Prompt: When the person touches the mirror, make the mirror ripple beautifully like liquid, and the person's arm turns into reflective mirror material

Prompt: When the person touches the mirror, the person transforms into a detailed monochrome line art drawing

Prompt: When the person touches the mirror, the person suddenly transforms into a cute felted stuffed puppet version with large googley eyes and glasses

Prompt: When the person touches the mirror, the person instantly transform into a vintage monochrome transparent 3d line art hologram, inside of a monochrome 3d holodeck maintaining the structure and details of the room and environment

Prompt: When the person touches the mirror, the entire environment turns into 3d voxel art

Reimagine the action

Switch up what happens in your videos, from the ordinary to the spectacular.

Prompt: Make it look like the weird shape of my hand hole super zooms and magnifies the ground it's looking at in sharper quality.

Prompt: When the finger in

Prompt: The lights of the apartments start turning on in sync with the music.

Edit real videos based on images

Use reference images to edit your creations, giving you even more creative control.

Prompt: When the hand opens, make a vast 3d architectural structure based on this image start building upward, sitting in the palm of the hand, reflecting prismatic light onto the hand and table. It builds with a 3d wireframe holographic effect. No music, just realistic real world sound.

Prompt: When the hand opens, reveal a sun floating in the center of the hand (sun should be animated, subtle solar flare movement) with bronze balls orbiting around it in mid air (no wires). When the hand opens make the lights dim to become nighttime, but keep the video the same until the hand opens. No music, just realistic sound.

Prompt: When the hand opens, reveal a physical photorealistic flying machine based on this sketch, floating above the hand, propeller spinning. No music, just realistic sound.

Edit over multiple turns, with consistency

Craft your scene step-by-step, changing specific details, environments, camera angles, and more.

Input video

Prompt: Transport the violinist to the image environment

Prompt: Make the violin invisible

Prompt: Change the camera angle to be over the violinist’s shoulder.

Swap in different objects or characters with natural language

Replace characters and objects in your video just by asking, all while maintaining a coherent, cohesive scene.

Prompt: Change spaceship to

Bring ideas to life, grounded in Gemini’s world knowledge

Create scenes that follow real-world logic. Gemini Omni pulls from its deep knowledge of history, biology, and narrative logic to construct compelling stories.

Create output that follows real-world physics

Omni has an intuitive understanding of forces like gravity, kinetic energy, and fluid dynamics for more realistic movement.

Prompt: A marble rolling fast on a chain reaction style track, continuous smooth shot

Draw on real-world history, science, and math

Omni understands world history, science, and math – and knows how to craft stories around it.

Prompt: claymation explainer of protein folding, everything is made out of clay, no hands, stop motion, accurate

Prompt: A skeuomorphism stop motion explainer about how the brain hippocampus works with a compelling voiceover. Don’t add seahorses. No voice cuts at the end. Don’t add text.

Sync text with onscreen action

Go beyond just rendering realistic text. Create videos that coherently connect text to what’s happening in the video.

Prompt: The video shows items of the alphabet. An unusual item starting with each letter is shown sitting on a table (like a Capybara for C, disco globe for D and Lava Lamp for L). All 26 letters must be represented by 26 items with matching lower thirds displaying the letter. Only one item and lower third at a time. Each lower third must look like a black marker written on a slip of paper in the bottom left. Rapid fire, roughly 9 frames per item at 24FPS. Last frame is a slip of paper "THE END". The whole video is accompanied by calm smooth music.

Prompt: word by word, one word on a the screen at a time: did, you, know, that, this, model, can, do, pretty, good, text!? each word appears with a different animated style, perfect pacing to a rhythm, sizzle reel

Reference anything

Reference and combine ingredients to maintain control and consistency over your scene.

Combine multiple inputs

Prompt with different inputs, and leave Gemini Omni to craft them into a single compelling narrative.

Prompt: refer to the extreme camera movement, perspective, distortion in

Prompt: Add harp sounds synchronized to when I touch each fern leaf. Change the leaf structure to all resemble semi translucent 3d bioluminescent plant life, with bioluminescent fireflies flying around it that react as I play, in sync with the sounds, subtle bokeh depth of field dynamic lighting, relecting off the walls in the room, keeping the room structure the same

Prompt: Imagine the world gradually changing into retro futuristic style (grainy and moody as ) as I walk. Use the audio for a retro-futuristic background music. 10s.

Transfer motion and styles

Apply motion and style references from an image or video across to your output.

Prompt: Apply the pose and motion from input video to provided character from this image. Apply style from image reference to the new video

Prompt: Rose is made from this crystal-like material

Prompt: Apply the motion of the whale swimming from the provided video to the provided image of fluid reflective material. Do not show the whale or water; instead, have this reflective moving material form a shape that resembles the whale as it swims. Replace water with white smooth material shapes that move

Swap characters or objects with a reference image

Provide an image of a character with your video, and the new character will match your motion and dialogue seamlessly.

Prompt: turn me into this character

Prompt: turn me into this character

Prompt: turn me into this character

Translate drawings into video

Turn sketches into realistic video – and use your doodles to guide how individual elements should move.

Prompt: turn this into realistic footage, using the drawing only as a guide for movement, do not show the drawing in the final video

Prompt: turn this into realistic footage, using the drawing only as a guide for movement, do not show the drawing in the final video

Prompt: turn this into realistic footage, using the drawing only as a guide for movement, do not show the drawing in the final video

Creating your prompts

Use our prompt guide to create realistic, coherent, and creative output.

Learn how to prompt

Safety

From development to deployment

Gemini Omni Flash was developed in partnership with internal safety, security, and responsibility teams. A range of evaluations and red teaming activities were conducted to help improve the model and inform decision-making. These evaluations and activities align with Google's AI Principles and responsible AI approach, as well as Google's Generative AI policies (e.g. Gen AI Prohibited Use Policy and the Gemini API Additional Terms of Service). Evaluation types included but were not limited to:

Training/development evaluations including automated and human evaluations carried out continuously throughout and after the model’s training, to monitor its progress and performance

Human red teaming conducted by specialist teams who sit outside of the model development team, across the policies and desiderata, deliberately trying to spot weaknesses and ensure the model adheres to safety policies and desired outcomes

Automated red teaming to dynamically evaluate Gemini Omni Flash for safety and security considerations at scale, complementing human red teaming and static evaluations

Ethics and safety reviews conducted ahead of the model’s release

Content created or edited with Omni in the Gemini app, Google Flow or YouTube includes our imperceptible SynthID digital watermark and C2PA Content Credentials. You can easily verify content through the Gemini app and coming soon to Chrome and Search. You can find out more about how we're expanding our content transparency and verification tools to help you understand how content was created and edited across the web in our blog post.

Learn more

Try Gemini Omni

Google AI subscription required. Features vary by tier and geography.

联系我们 contact @ memedata.com