Amar Singh

4.7K posts

Amar Singh

Amar Singh

@ummerrr

generative media PM at Google Flow (@flowbygoogle), previously google shopping. opinions are my own

Brooklyn, NY Katılım Nisan 2010
949 Takip Edilen455 Takipçiler
Gavin Purcell
Gavin Purcell@gavinpurcell·
my cat didn't tell me she was starting a cult (aka good lord, google omni editing of irl footage is remarkable)
English
1
0
23
752
Amar Singh
Amar Singh@ummerrr·
@alisa_fortin It's such a unique and fun model. It's been hard to fully articulate, but it keeps surprising me
English
1
0
1
16
Amar Singh
Amar Singh@ummerrr·
@DeryaTR_ We've been having a ton of fun internally. Character consistency is also really impressive with Gemini Omni
English
0
0
1
47
KARIM
KARIM@sat0oshi·
Google OMNI le nouveau modèle vidéo de google est disponible sur FLOW la plateforme de génération d'image de google. @FlowbyGoogle on va tester ça hein !
KARIM tweet media
Français
1
0
4
2.4K
Amar Singh retweetledi
Google
Google@Google·
Since launching @FlowByGoogle last year, millions of creatives have used it to bring their ideas to life. Today we’re rolling out three new updates for creative work on Google Flow: - Gemini Omni Flash - Google Flow Agent - Google Flow Tools #GoogleIO
English
14
35
220
38.5K
Amar Singh
Amar Singh@ummerrr·
People have been absolutely loving characters in Google Flow 🌊
Google@Google

Now in @FlowByGoogle, you’ll be able to use the full potential of Gemini Omni Flash, our model that can create anything from any input, starting with video. With a simple prompt and style reference, Gemini Omni allows you to transform the environment of an existing scene, add visual effects and other elements, all while preserving the original performance. You can even add new characters with custom voices directly to a scene with the new Character feature. #GoogleIO

English
0
0
0
62
David Pantera
David Pantera@davidpantera_·
Wow Omni really is a nano banana moment for video. Prompt: Create a video of the major event that happened at these coordinates: 42°21'15.1"N 71°03'07.9"W
English
4
1
24
1.9K
Ethan Mollick
Ethan Mollick@emollick·
I think people don't realize why Gemini Omni is different than other video AIs. It is fully multimodal, so it can edit video natively, too I took the famous "train " movie from 1896 & made it a bullet train, LEGO, added a time traveler, a centipede, muppets... (see reflections?)
English
107
184
2.3K
177.6K
Amar Singh
Amar Singh@ummerrr·
@bmx_ai13 I have made many, many dog on skateboard video with Gemini Omni 🐶🛹 It's been really satisfying to swap with different animals and see how Omni reacts
English
0
0
0
20
BMX
BMX@bmx_ai13·
Gemini Omni Test
Eesti
5
3
23
748
Amar Singh
Amar Singh@ummerrr·
@MrDavids1 We appreciate you checking out the new experience and we really appreciate the thoughtful feedback! Your feedback is good and fair - we are listening and still working hard to improve. To that end, tagging some folks on our team for visibility @nmatares @oddclaw @ronnyvotel
English
2
0
2
50
Travis Davids
Travis Davids@MrDavids1·
Ok I'm going to be brutally honest, I've been testing the new updates in Google Flow. There is just nothing innovative that was released. Why has Google reached a point of stagnation? I'm honestly surprised. Seedance 2.0 is still winning this race. I'm just going to throw some random thoughts out there from testing, highlighting both pro's and cons. This is honest feedback to @googleaidevs 🟩Pro's - Being able to attach a consistent voice to a character is actually a great new improvement. This needs to be the standard with all AI video generators. I don't know why it took so long to implement this feature BUT I'm happy it exists. - Being able to create, save and easily reference a character is also a nice addition. - Yes, the Agent is a nice added touch as well. It can come in handy for brainstorming ideas. 🟥Cons - 10 second generations feel outdated, I thought Google would have tried to push for 15 for a single generation, or maybe we'll get that with Veo 4? - Ok, you can do edits in a video. Cool but...there is nothing innovative here. We've been able to do this with Pika, Runway Aleph, Kling, Seedance 2.0 for a long time now. It's offered nothing new. Also, 10 seconds is limiting with video edits as well. - Not being able to extend a Google OMNI generated video makes no sense to me. Why does this limitation even exist? By the way, anyone else having issues with using character sheets? Overall, this update was...meh. It feels like the needle didn't move much. What does everyone else think?
English
3
0
5
339
Anna writes
Anna writes@anna_writes0·
@alwaysmoses12 No, I haven’t yet. I’ll check it out. Is it mainly used for image generation?
English
1
0
0
15
Anna writes
Anna writes@anna_writes0·
I hate to admit it, but Google Flow AI is honestly one of the best tools for image generation right now ❤️‍🔥
English
8
4
40
1.3K
Dheepan Ratnam
Dheepan Ratnam@Dheepanratnam·
Google Omni in Flow 🧵 just dropped a massive glow-up. check how its made in the comments Same character. Same voice. Consistent across completely different scenes. The consistency on face, scars, voice, and vibe is wild. This is Elias Thresher business guy + yacht owner generated in Flow, then dropped into the scene builder. First generation and it’s already this clean. Watch him living his best life in Greece:
English
11
13
115
7.9K
Ben Nash
Ben Nash@bennash·
Here's a nice intro to Google Omni (the best product update from Google this week). IMO it is the "Nano Banana" of video. Also Google Flow is much better. Google OMNI Does What Every AI Creator Has Been Waiting For... youtu.be/LKig4_qePxo?si… via @YouTube
YouTube video
YouTube
English
1
0
10
895
Anna writes
Anna writes@anna_writes0·
@martinhurley 😂 Honestly, I was surprised, too. The image quality is crazy good especially the realism, lighting, and cinematic style. I'm still testing it more, but it’s definitely one of the strongest AI image tools I’ve used recently.
English
2
0
1
91