Mion.

thoughts (in progress)
TIP

[Mion. @mionwang]

@StrayRollie @pubity Well not really. First of all, they're still trying to crack agi with their flagship data centers.

Musk just saw that even if AI tech hits a wall and progress stagnates, the world will still need a lot of compute for inference. Plus they're gonna put data centers in space.

[Mion. @mionwang]

@IndianTechGuide Say goodbye to all the rubber seals

[80 LEVEL @80Level]

This Blender simulation created by MaxMilez looks so slimy and sticky that you can even imagine how it sounds.

It was made using an add-on, and it will be available to the public: https://t.co/w4aTOIP15N

[UI/UX Savior @UiSavior]

🤥

[@jason @Jason]

Business AI is here.. for your business! 😂

[Ideogram @ideogram_ai]

Introducing Ideogram 4.0: the best open image model in the world.

Think it. Make it. Own it.

Download the weights, fine-tune on your own data, and run it on your hardware. Live on every Ideogram plan and the API today.

[Jerrod Lew @jerrod_lew]

Ideogram 4.0 has arrived.

The latest text-to-image AI image model comes with higher quality, text generation and is also open weights.

Here's everything you need to know.

[Mion. @mionwang]

@OpenAI i got excited and thought this video was ai generated for a sec... wait is it?

[Mion. @mionwang]

@mhdfaran It's impressive and trash at the same time.. Crazy

[patrick. @imPatrickT]

this is what real filmmaking looks like and nothing will ever replace it

[Runway @runwayml]

Aleph 2.0 is now available via the Runway API. Bring precise video editing directly into your apps, products and platforms. Edit up to 30 seconds of video at 1080p across multi-shot sequences, changing only what you want.

Get started at the link below.

[Blackmagic Design @Blackmagic_News]

DaVinci Resolve 21 final release now available! Major update adds hundreds of features including new AI tools such as IntelliSearch for fast content searching and CineFocus for focal point adjustment! Plus new motion graphic effects, Fairlight folder tracks and the Photo page for still images! Download now from https://t.co/MNXwybBK2b

[Mion. @mionwang]

I.e. it must be abstracted from human language and it needs to have an understanding of time somehow.

[Mion. @mionwang]

I do think that JEPA like neutral nets is the future. But I don't know if Yann LeCun's JEPA is it - mainly because I don't understand any of it.

But I know for a fact that LLMs aren't going to be enough to solve true human like general intelligence.

[Mion. @mionwang]

@chrisfirst @higgsfield Was it absolute dogshit or just shit?

[Nikita Bier @nikitabier]

Headed home now. Here’s a quick recap of my Japan trip.

[Bilawal Sidhu @bilawalsidhu]

Camera path scribbles are turning into an AI video trend. Cool seedance example by Keety.

[Mion. @mionwang]

@tibininin Ah

[Aparajite @amshilparaghu]

⚠️ Bought Whole Farm roasted Channa from Blinkit. Even after washing it multiple times, it’s still leaving yellow colour in the water 😠

Saw a page on insta - unfiltered.haribros video highlighting the same issue. Is this artificial dye or something? Not feeling safe consuming it.

@letsblinkit please clarify or take action on quality!

#WholeFarm #Blinkit #Chana #FoodSafety

This is supposed to be a healthy snack, not a chemical one. Anyone else faced this?

[fofr @fofrAI]

One year later with Omni and this test can pass. I saw it getting pretty close, so I tweaked the prompt:

> A video of a man counting to 10 on his fingers, show the number in the corner. A new number every 1s, no dialogue other than the numbers he says. He uses two hands for numbers bigger than 5.

- the model does 1 to 5 consistently well
- struggles more when two hands are used, usually on 7 and 8
- if you ask it to count faster, errors increase
- it keeps a good cadence