Users can create reusable avatars from short recorded clips.
Omni from @GoogleDeepMind just dropped 👀
It's a big step forward in video generation when it comes to character consistency, world knowledge, and editing.
I've been testing it for the last few days - and I'm excited to walk through some of the key features + my clips 👇
Positive users praise Google DeepMind's Omni for realistic AR video filters, precise edits like mirror reflections, and strong prompt-based results, while negative users criticize frequent failures, errors, and frustrating unreliability.
i don’t think you understand how insane omni is
Amazing work by Google once again 😍
@GoogleDeepMind 1) Avatars
Record a clip of yourself, and your face + voice will be saved as a character that you can add to any video.
This makes it so easy to put yourself in any clip, while changing your style or outfit as needed to adapt to the scene.
These clips are all my avatar 🤯
Here's an example of 3 edits of a video with Omni:
1. original 2 maker her invisible, put gloves on her 3. while she's talking, two men come and take away the framed picture 4. change her outfit
@GoogleDeepMind 1) Avatars
Record a clip of yourself, and your face + voice will be saved as a character that you can add to any video.
This makes it so easy to put yourself in any clip, while changing your style or outfit as needed to adapt to the scene.
These clips are all my avatar 🤯
Omni from @GoogleDeepMind just dropped 👀
It's a big step forward in video generation when it comes to character consistency, world knowledge, and editing.
I've been testing it for the last few days - and I'm excited to walk through some of the key features + my clips 👇
> Make the whole wall behind her a mirror
Note the camera position in the reflection.
Here's an example of 3 edits of a video with Omni:
1. original 2 maker her invisible, put gloves on her 3. while she's talking, two men come and take away the framed picture 4. change her outfit
Omni isn't Veo/Seddance!
I had the awesome pleasure to have breakfast with @JeffDean during #googleio and ask him about Omni and how it differs from "video models" in the classic sense like Veo, Seedance and HappyHorse
Huge thanks to the deepmind team for the invite!
Omni is pretty nuts. It is NOT seedance. Any input in/out. It’s more than nano banana for video - it’s quite literally industrial light & magic. Now effectively reduced to an insanely realistic AR filter that you can apply on demand to anyone’s footage.
Omni video to video 👌
Transforming materials with Gemini Omni. Prompt + video input in 🧵
Omni is pretty nuts. It is NOT seedance. Any input in/out. It’s more than nano banana for video - it’s quite literally industrial light & magic. Now effectively reduced to an insanely realistic AR filter that you can apply on demand to anyone’s footage.
> Redo this video so that there is a new person every 1s (or 24 frames at 24fps). Rapid fire.
Here's an example of 3 edits of a video with Omni:
1. original 2 maker her invisible, put gloves on her 3. while she's talking, two men come and take away the framed picture 4. change her outfit
There's a new Release Notes podcast dropping soon, it'll be a really nice deep dive into Omni with the folks that helped make it.
But also, I had far too much fun using Omni to make this preview.
2) World knowledge
Omni is grounded in Gemini's world knowledge - which means that it just knows a LOT of things without you having to include it in the prompt.
For example, upload an image of where you're standing and ask for a history or to explain a topic (like healing an ACL tear).
@GoogleDeepMind 1) Avatars
Record a clip of yourself, and your face + voice will be saved as a character that you can add to any video.
This makes it so easy to put yourself in any clip, while changing your style or outfit as needed to adapt to the scene.
These clips are all my avatar 🤯
One of the full Omni edits from this intro. It's wild to me how natural the monkey is, and how it decides to play with his glasses (unprompted).
There's a new Release Notes podcast dropping soon, it'll be a really nice deep dive into Omni with the folks that helped make it.
But also, I had far too much fun using Omni to make this preview.
> Rapid fire but consistent. No dialogue. Just a simple music track. Each shot takes 1s or 24 frames at 24FPS. Captivating cinematic music. Theme is a high budget trailer for a nature documentary, include timelapses. A voice says: "This is Omni. The new model from Google"
Now you can just put cats on people's heads
Here's an example of 3 edits of a video with Omni:
1. original 2 maker her invisible, put gloves on her 3. while she's talking, two men come and take away the framed picture 4. change her outfit
What the hell is going on?
i don’t think you understand how insane omni is
So good @OfficialLoganK 😂
i don’t think you understand how insane omni is
@GoogleDeepMind 5) Multi-image prompting
Omni can take up to five images and one video as a prompt.
I've been putting this to the limit - taking screenshots of Zillow listings and dumping them into the model.
I've been pretty impressed with the results! (and want >10 seconds 😉)
what they feed us in the Gemini microkitchens!
By default Omni has a bit of a professional "look", but can also create "normie" videos with the right prompting.
@GoogleDeepMind That's it for now! I'm sure I'll be back with more generations later 😅
Huge congrats to @shlomifruchter, @doomie, @nbrichtova + the team on the launch.
Really excited to see what folks do with "Nano Banana for video."
@GoogleDeepMind 3) Video editing
You can upload real videos to Omni and ask for edits - changing the action, style, or subject. Or you can annotate on top of a video.
In this video, I asked to "change my hat every time I clap."
