wave3d.ai

Wave3d is a spatial design and creative language for seamless collaboration between human and ai agents

[wave3d.ai @DanielPPFW]

@chetanankola @ElevenLabs @clemmygames @threejs @OpenAI wow the background blur in real time as camera moves is just too good

[wave3d.ai @DanielPPFW]

Credits
Music: "Medieval: Harvest Season" by RandomMind (CC0), https://t.co/3k8jGFbl8M
Alpaca model: @quaternius, via Poly Pizza, https://t.co/0ZWd2oZDkT

[wave3d.ai @DanielPPFW]

Last time, 5,000 particles in live force fields. This time, over 1,000,000 strands.

Two ranch alpacas, a mother and her cub, in a meadow where every strand of fur and every blade of grass answers the same live wind, in real time.

Built by AI agents on Wave Engine, our framework on top of @threejs. The strand count isn't the story. How little the agents had to write, and how little they had to figure out, is. To capture that, we coined two measures of our own:

FLTC, framework-level token compression: far less code to write.
FLRC, framework-level reasoning compression: far less to reason through.

Details below.

[wave3d.ai @DanielPPFW]

The numbers: about 2,000 lines of Wave code for the whole scene, with every model and material generated in code except the imported alpaca. By our estimate, roughly 80% less reasoning (FLRC) than the same scene in raw three.js.

Most tools buy savings like that by packaging features for one genre. It works until you step outside the genre.

Wave has no fur system and no alpaca feature. One wind moves the fur, the grass and the weathervane; the same camera direction would shoot any subject. The agents composed this from general building blocks, and this demo uses a small slice of them: thousands of generalizations, refined through years of accumulated experience and reinforcement learning.

That is why it compounds. Every scene an agent builds on Wave starts with that experience behind it.

More soon.
#Wave3DAI #threejs #Web3D #WebGPU

[wave3d.ai @DanielPPFW]

@elonmusk @minchoi crazy ramp up

[wave3d.ai @DanielPPFW]

@IndependentEco SPACEX rocks

[wave3d.ai @DanielPPFW]

@fecolinhares @lyraxana i can not agree more with this

[wave3d.ai @DanielPPFW]

@gruberbuilds @threejs @ElevenLabs @tripoai i think opus5.5 can model the seagull procedurally too!

[wave3d.ai @DanielPPFW]

@cryptomanavan ask opus 5.5 to migrate the rendering pipeline to webGPU it might just do the trick for you

[wave3d.ai @DanielPPFW]

@pcshipp i just did

[wave3d.ai @DanielPPFW]

@cryptomanavan not bad - 30~40 fps on my 3080ti laptop. i suppose it's rendered with webGPU ?

[wave3d.ai @DanielPPFW]

@loktar00 imagine 1 year from now, deepseek 7.1 flash will do the things opus5.5 can do today.. at the cost of deepseek 4.1 flash today.

[wave3d.ai @DanielPPFW]

@dangreenheck Honestly the most amazing thing here is the level of optimization that is done to make this 60FPS even on my M4 macbook air for everything, with no fog at distance. I climbed to the top of the mountain and looked down. wow. no culling everything is there to see

[wave3d.ai @DanielPPFW]

@loktar00 imagine what fable5.5 can do. i think it's already there, Ant just chose not to release it because opus5.5 can single-handedly beat OAI

[wave3d.ai @DanielPPFW]

@BrandonVanB @thsottiaux @OpenAI getting this a lot in the fast two hours.

[wave3d.ai @DanielPPFW]

@dangreenheck Your water pro is still the best in class when detail matters. Sadly AI has learned to copy every it sees with 90% similarity - but that 10% still feels different