This PDoom video made by Claude is going super viral major AI researchers are interacting with it
The dev just open sourced it sending creator rev to their Git
https://t.co/frR525yTVO
Okay hear me out I'm literally tokenizing the first four dimensional dog onchain and maybe the first ever true four dimensional coin in general
How it works in simple terms:
1. Start with a 2D video of a shiba running in the grass. This becomes the height and width of the cube (xy)
2. Pull the video apart into frames
3. Stack the frames along the third axis of the cube. This axis is time (t)
Jev-style logit read on a 4B open VLM, measured: https://t.co/bvA3ijRgkU 👀
vs the same model writing JSON:
⏳ ~1/3 less time for a yes/no on a full photo
🤑 up to ~85% less GPU cost with many Qs per image
🎯 same accuracy
on fresh photos: level with the best hosted models on pick-one (0.933 vs 0.937), ~2 pts behind Gemini on yes/no
pip install glance-vlm
https://t.co/rtdSOPY3pD
We're running Grokification becasue some random artist made Grok Bot memes
While we have a Meta employee literally encouraging us to make muse memes of people
Muse is way more viral than Grok Bot right now
Is this not the perfect Museification setup??
if you too want to make adorable little muse pics, I just give my muse this photo and prompt it with “make a photo of this guy <insert the blank>”
i am unleashing all of twitter to make adorable muse pics using muse. don’t let me down!