Dagmawi Babi
4.7K subscribers
12.7K photos
1.58K videos
238 files
1.66K links
Believer of Christ | Creative Developer.

Files Channel: https://t.me/+OZ9Ul_rSBAQ0MjNk

Community: @DagmawiBabiChat
Download Telegram
Dagmawi Babi
Atleast I got early access to Gemini 1.5 and Google AI Studio :) #Gemini #AI #Google @Dagmawi_Babi
Google AI Studio and Gemini 1.5 Pro Review

• 1Million tokens is impressive out of the box
• It can analyze full files, video and pictures well and can be queried
• You can fine-tune the safety, it has built in API access and code generation is better
• But the studio is janky, inefficient, and buggy
• Got bored of it quickly

#Google #Gemini #AI
@Dagmawi_Babi
This media is not supported in your browser
VIEW IN TELEGRAM
Google DeepMind just dropped 'Genie', an AI that can generate interactive video games.

This is a huge deal.

Genie is trained on 200,000 hours of unsupervised public internet gaming videos and can generate video games from a single prompt or image.

But here's what's insane:

Despite not being trained on action or text annotations, the foundation model can determine who the main character is and enable a user to control that character in the generated world.

It does this through its Latent Action Model, Video Tokenizer, and Dynamics Model (will go more in-depth on this in tomorrow's newsletter for those interested).

And for those asking, yes, it's research-only and not publicly available (here come the Google memes), and it does come with some limitations, like only currently creating games at 1FPS.

But this is the worst AI will ever be.

#Google #Genie #AI
@Dagmawi_Babi
Dagmawi Babi
Google DeepMind just dropped 'Genie', an AI that can generate interactive video games. This is a huge deal. Genie is trained on 200,000 hours of unsupervised public internet gaming videos and can generate video games from a single prompt or image. But here's…
It's actually kinda blowing my mind thinking about the engineering.

Genie enables users to act in the generated environments on a frame-by-frame basis despite training without any ground-truth action labels or other domain-specific requirements typically found in the world model literature. Further the resulting learned latent action space facilitates training agents to imitate behaviors from unseen videos.

Let me simply explain this:
You can show it a picture and it can separate the world from the character AND define its own world logic frame by frame.

You literally don't have to put collider boxes on game items for them to interact, it just does that without that.

Literally u don't have to define the logics of the game world you're building, u can if u want to, but u don't have to.

This's oddly a super freaking interesting idea.

#Google #Genie #AI
@Dagmawi_Babi
And even more simply put...

You can easily generate an entire video game based on a sketch or a picture u have. The AI will handle literally everything else including expanding the world, character control, game logic open world interactions and so much more! 🤯

#Google #Genie #AI
@Dagmawi_Babi
Here is the research paper
https://t.me/c/1156511084/813

Let's not even talk about how they trained it without human supervision on unlabelled videos. AND it scaled well as they increased GPUs which most LLMs don't.

#Google #Genie #AI #Papers
@Dagmawi_Babi
This's Gemini as a personal assistant on the latest version of Android. It's pointless cause it can't control the phone like Google Assistant.

#Gemini #Google
@Dagmawi_Babi
Finally got my hands on the Google Pixel 7. It's a lovely compact, good UX phone. Defo my type.

#Google #Pixel7
@Dagmawi_Babi
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
Media is too big
VIEW IN TELEGRAM
Project Astra, it's really impressive!
And looks so nice to use. 🔥

#Google #ProjectAstra
@Dagmawi_Babi
ChatGPT now allows free users to generate 2 DALL-E images per day. (Lol cheap asses)

But most importantly google has lowered their Gemini Flash API pricing from $0.35 to $0.0075 which's insane 🤯

#Google #OpenAI #ChatGPT
#DALLE #Gemini #AIML
@Dagmawi_Babi
I actually enjoyed watching Google's Made by Google event.
https://www.youtube.com/watch?v=N_y2tP9of8A

I also think the Pixel 9 phone is gorgeous and very powerful cause almost all gen AI is running on device.

#Google #TechEvents
#MadeByGoogle #Pixel9
@Dagmawi_Babi
Google's new image generator is insanely good! 🔥
https://aitestkitchen.withgoogle.com/tools/image-fx

It's currently free for a while.

#Google #AIArt #Imagen3
@Dagmawi_Babi
Hopefully a lot more phones here connect

#Google
@Dagmawi_Babi
Also I am in love with Gemini Flash. It's soooo fast and compact yet so powerful it's so impressive.

Not to mention it's so cheap I could afford it.

#Google #Gemini #LLM #AIML
@Dagmawi_Babi